Construction method and device of three-dimensional model, equipment, medium and product

Abstract

The embodiment of the invention provides a three-dimensional model construction method and device, equipment, a medium and a product. The method comprises the following steps: acquiring a light path, and splitting the light path to obtain at least one sub-path; for any sub-path, carrying out micro-mapping on any second light ray direction in the second integral domain, and determining a first light ray direction corresponding to the second light ray direction from a plurality of light ray directions in the first integral domain; the first integral domain is an effective integral domain of a coloring point corresponding to the sub-path before the scene parameter is updated; the second integral domain is an effective integral domain of scene parameter updating and coloring points; constructing differential homeomorphism corresponding to the sub-paths; the differential homeomorphism indicates the corresponding relation between the first light direction and the second light direction; re-parameterizing an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path to obtain a re-parameterization integral function; and constructing a target three-dimensional model according to the re-parameterization integral function. And the construction efficiency of the target three-dimensional model is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of computer graphics, and particularly relate to a three-dimensional model construction method, device, equipment, medium and product. BACKGROUND

[0002] Differentiable rendering is a technology of reconstructing a three-dimensional model of an object or a scene based on a two-dimensional image, and is a hot research content in the field of computer graphics. In the construction process of the three-dimensional model, there is often an occlusion relationship between geometric objects in the scene, thereby causing the integral domain of the path integral function to be discontinuous.

[0003] In the related art, the path integral function can be re-parameterized, and based on the re-parameterized path integral function, a target three-dimensional model is constructed.

[0004] However, in the related art, the information required in the re-parameterization process is usually determined by means of auxiliary light and ray tracing, thereby causing the method in the related art to have a long time-consuming process of re-parameterizing the path integral function, and further causing the method in the related art to have a low construction efficiency of the three-dimensional model. SUMMARY

[0005] Embodiments of the present application provide a three-dimensional model construction method, device, equipment, medium and product, which improve the construction efficiency of the three-dimensional model.

[0006] In a first aspect, embodiments of the present application provide a three-dimensional model construction method, comprising:

[0007] obtaining a light path and splitting the light path to obtain at least one sub-path;

[0008] For any sub-path, differentiable mapping is performed on any second light direction in a second integral domain to determine a first light direction corresponding to the second light direction from a plurality of light directions in a first integral domain; the first integral domain is an effective integral domain of a shading point corresponding to the sub-path before updating of scene parameters; and the second integral domain is an effective integral domain of the shading point after updating of the scene parameters;

[0009] constructing a differential homeomorphism corresponding to the sub-path; the differential homeomorphism indicates a corresponding relationship between the first light direction and the second light direction;

[0010] performing re-parameterization processing on an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path to obtain a re-parameterized integral function;

[0011] constructing a target three-dimensional model according to the re-parameterized integral function.

[0012] In an implementation manner, the differentiable mapping is performed on any second light direction in the second integral domain, and the first light direction corresponding to the second light direction is determined from the plurality of light directions in the first integral domain.

[0013] The first segmentation unit and the second segmentation unit corresponding to the sub-path are determined; the first segmentation unit belongs to the target three-dimensional object before the scene parameter update, the second segmentation unit belongs to the target three-dimensional object after the scene parameter update, and the target three-dimensional object corresponds to the sub-path;

[0014] The position information of the second mapping point to which the second light direction is mapped to the second segmentation unit is determined;

[0015] The position information of the second centroid point of the second segmentation unit is determined, and the position information of the first centroid point of the first segmentation unit is determined;

[0016] The position information of the first mapping point is determined according to the position information of the second mapping point, the position information of the second centroid point and the position information of the first centroid point;

[0017] The first light direction is determined from the plurality of light directions included in the first integral domain according to the position information of the first mapping point; the first mapping point is a mapping point of the first light direction mapped to the first segmentation unit.

[0018] In an implementation manner, the position information of the first mapping point is determined according to the position information of the second mapping point, the position information of the second centroid point and the position information of the first centroid point, including:

[0019] The relative position is determined according to the position information of the second mapping point and the position information of the second centroid point;

[0020] The position information of the first mapping point is determined according to the relative position and the position information of the first centroid point.

[0021] In an implementation manner, the integral function corresponding to the light path is re-parameterized according to the differential homeomorphism corresponding to at least one sub-path to obtain a re-parameterized integral function corresponding to the light path, including:

[0022] The Jacobian corresponding to the sub-path is constructed according to the first segmentation unit and the second segmentation unit;

[0023] The integral function corresponding to the light path is re-parameterized according to the differential homeomorphism corresponding to at least one sub-path and the Jacobian corresponding to at least one sub-path to obtain a re-parameterized integral function corresponding to the light path.

[0024] In an implementation manner, the Jacobian corresponding to the sub-path is constructed according to the first segmentation unit and the second segmentation unit, including:

[0025] determine whether the first integral domain and the second integral domain belong to a planar integral domain;

[0026] In a case where the first integral domain and the second integral domain do not belong to the planar integral domain, determine the Jacobian corresponding to the sub-path according to the acquired position information of the first segmentation unit, the acquired position information of the second segmentation unit, the position information of the first mapping point and the position information of the second mapping point.

[0027] In an implementation manner, the method further includes:

[0028] In a case where the first integral domain and the second integral domain belong to the planar integral domain, project the first segmentation unit onto a preset image plane to obtain a first projection unit;

[0029] project the second segmentation unit onto the preset image plane to obtain a second projection unit;

[0030] determine the Jacobian corresponding to the sub-path according to the acquired position information of the first projection unit and the acquired position information of the second projection unit.

[0031] In a second aspect, an embodiment of the present application provides a three-dimensional model construction device, including:

[0032] a processing module, configured to acquire a light path and split the light path to obtain at least one sub-path;

[0033] The processing module is further configured to, for any sub-path, perform differentiable mapping on any second light direction in a second integral domain, and determine a first light direction corresponding to the second light direction from a plurality of light directions in a first integral domain; the first integral domain is an effective integral domain of a shading point corresponding to the sub-path before scene parameter updating; and the second integral domain is an effective integral domain of the shading point after scene parameter updating.

[0034] The processing module is further configured to construct a differential homeomorphism corresponding to the sub-path; the differential homeomorphism indicates a corresponding relationship between the first light direction and the second light direction.

[0035] The processing module is further configured to perform reparameterization processing on an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path, to obtain a reparameterized integral function.

[0036] a construction module, configured to construct a target three-dimensional model according to the reparameterized integral function.

[0037] In an implementation manner, the processing module is specifically configured to include:

[0038] determine a first segmentation unit and a second segmentation unit corresponding to the sub-path; the first segmentation unit belongs to the target three-dimensional object before the scene parameter update, and the second segmentation unit belongs to the target three-dimensional object after the scene parameter update, and the target three-dimensional object corresponds to the sub-path;

[0039] determine position information of a second mapping point of the second light direction mapping to the second segmentation unit;

[0040] determine position information of a second centroid point of the second segmentation unit, and determine position information of a first centroid point of the first segmentation unit;

[0041] determine position information of the first mapping point according to the position information of the second mapping point, the position information of the second centroid point and the position information of the first centroid point;

[0042] determine the first light direction from a plurality of light directions included in the first integral domain according to the position information of the first mapping point; the first mapping point is a mapping point of the first light direction mapping to the first segmentation unit.

[0043] In an implementation manner, the processing module is specifically configured to:

[0044] determine a relative position according to the position information of the second mapping point and the position information of the second centroid point;

[0045] determine the position information of the first mapping point according to the relative position and the position information of the first centroid point.

[0046] In an implementation manner, the processing module is specifically configured to:

[0047] construct a Jacobian corresponding to the sub-path according to the first segmentation unit and the second segmentation unit;

[0048] perform reparameterization processing on the integral function corresponding to the light path according to the differential homeomorphism corresponding to at least one sub-path and the Jacobian corresponding to at least one sub-path, to obtain a reparameterization integral function corresponding to the light path.

[0049] In an implementation manner, the processing module is specifically configured to:

[0050] determine whether the first integral domain and the second integral domain belong to a planar integral domain;

[0051] in a case where the first integral domain and the second integral domain do not belong to the planar integral domain, determine the Jacobian corresponding to the sub-path according to the obtained position information of the first segmentation unit, the obtained position information of the second segmentation unit, the position information of the first mapping point and the position information of the second mapping point.

[0052] In an implementation manner, the processing module is further configured to:

[0053] In a case where the first integral domain and the second integral domain belong to a planar integral domain, the first split unit is projected onto a preset image plane to obtain a first projection unit;

[0054] The second split unit is projected onto the preset image plane to obtain a second projection unit;

[0055] According to the position information of the first projection unit and the position information of the second projection unit, a Jacobian corresponding to the sub-path is determined.

[0056] In a third aspect, an electronic device is provided, including a memory and a processor.

[0057] The memory stores computer execution instructions.

[0058] The processor executes the computer execution instructions stored in the memory, so that the processor executes the method of the first aspect.

[0059] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores computer execution instructions. When the computer execution instructions are executed by a processor, the computer execution instructions are used to implement the method of the first aspect.

[0060] In a fifth aspect, a computer program product is provided, including computer execution instructions. When the computer execution instructions are executed by a processor, the computer execution instructions make the method of the first aspect be executed.

[0061] Embodiments of the present application provide a three-dimensional model construction method, device, equipment, medium and product. In the method, an electronic device can obtain a light path and split the light path to obtain at least one sub-path. For any sub-path, the electronic device can perform differentiable mapping on any second light direction in a second integral domain (an effective integral domain of a shading point after updating of scene parameters), and determine a first light direction corresponding to the second light direction from a plurality of light directions in a first integral domain (an effective integral domain of a shading point before updating of scene parameters). The shading point corresponds to the sub-path. The electronic device can construct a differential homeomorphism corresponding to the sub-path. The differential homeomorphism indicates a corresponding relationship between the first light direction and the second light direction. The electronic device can perform reparameterization processing on an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path, to obtain a reparameterized integral function. The electronic device can construct a target three-dimensional model according to the reparameterized integral function. In this way, time-consuming ray tracing is avoided, the reparameterization rate of the integral function is improved, and the construction efficiency of the target three-dimensional model is improved. In addition, in this way, the variance of scene parameter gradient information obtained based on the reparameterized integral function is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced as follows. Obviously, the accompanying drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0063] Figure 1 A scene schematic diagram of a three-dimensional model construction method provided by an embodiment of the present application;

[0064] Figure 2a A flowchart of the first embodiment of the three-dimensional model construction method provided by an embodiment of the present application;

[0065] Figure 2b A schematic diagram of a light path provided by an embodiment of the present application;

[0066] Figure 2c A scene schematic diagram of a single-triangle unit provided by an embodiment of the present application;

[0067] Figure 2d A scene schematic diagram of a double-triangle unit provided by an embodiment of the present application;

[0068] Figure 2e A schematic diagram of multiple segmentation modes provided by an embodiment of the present application;

[0069] Figure 2f A schematic diagram of a corresponding relationship of a segmented integral domain provided by an embodiment of the present application;

[0070] Figure 2g A scene schematic diagram of a multiple-triangle unit provided by an embodiment of the present application;

[0071] Figure 3 A flowchart of the second embodiment of the three-dimensional model construction method provided by an embodiment of the present application;

[0072] Figure 4 A flowchart of the third embodiment of the three-dimensional model construction method provided by an embodiment of the present application;

[0073] Figure 5 A structural schematic diagram of a three-dimensional model construction device provided by an embodiment of the present application;

[0074] Figure 6 A structural diagram of an electronic device provided by the present application. DETAILED DESCRIPTION

[0075] To make the purposes, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments made by those of ordinary skill in the art under the inspiration of the embodiments of the present application belong to the scope of protection of the present application.

[0076] The terms "first", "second", "third", "fourth" and the like in the description, claims, and drawings of the embodiments of the present application (if any) are used to distinguish similar objects, and do not necessarily have to be used to describe a particular sequential or chronological order. It should be understood that the data thus used can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a list of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0077] Differentiable rendering is a technology for reconstructing a three-dimensional model of an object or a scene based on a two-dimensional image, and is a hot research content in the field of computer graphics. In the construction process of the three-dimensional model, there is often an occlusion relationship between geometric objects in the scene, resulting in a discontinuous integral domain of a path integral function.

[0078] In the related art, the path integral function can be reparameterized, and based on the reparameterized path integral function, a target three-dimensional model is constructed.

[0079] However, in the related art, the information required in the reparameterization process is usually determined by means of auxiliary light and ray tracing, resulting in the case that the process of reparameterizing the path integral function in the related art is time-consuming, and further resulting in the problem that the construction efficiency of the three-dimensional model in the related art is low.

[0080] Based on this, the embodiment of the application provides a three-dimensional model construction method. An electronic device can obtain a light path, and performs splitting processing on the light path to obtain at least one sub-path. For any sub-path, the electronic device can perform differentiable mapping on any second light direction in a second integral domain (an effective integral domain of a shading point corresponding to the sub-path after scene parameter updating), and determine a first light direction corresponding to the second light direction from a plurality of light directions in a first integral domain (an effective integral domain of the shading point before scene parameter updating). The electronic device can construct a differential homeomorphism corresponding to the sub-path, wherein the differential homeomorphism indicates a corresponding relationship between the first light direction and the second light direction. The electronic device can perform reparameterization processing on an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path to obtain a reparameterized integral function. The electronic device can construct a target three-dimensional model according to the reparameterized integral function.

[0081] In the above manner, the reparameterization rate of the integral function is improved, and the construction efficiency of the target three-dimensional model is further improved. In addition, in the above manner, the variance of the scene parameter gradient information obtained based on the reparameterized integral function is reduced.

[0082] The three-dimensional model construction method of the embodiment of the application is described in detail below.

[0083] Exemplarily, Figure 1 A scene diagram of the three-dimensional model construction method provided by the embodiment of the application is shown.

[0084] As Figure 1 shown, the electronic device 10 can obtain a light path, and split the light path to obtain at least one sub-path. It can be understood that the electronic device 10 can obtain the light path in response to a model construction request.

[0085] For any sub-path, the electronic device 10 can perform differentiable mapping on any second light direction in a second integral domain, and determine a first light direction corresponding to the second light direction from a plurality of light directions in a first integral domain. The first integral domain is an effective integral domain of a shading point corresponding to the sub-path before scene parameter updating; and the second integral domain is an effective integral domain of the shading point after scene parameter updating.

[0086] The electronic device 10 can construct a differential homeomorphism corresponding to the sub-path. The differential homeomorphism indicates a corresponding relationship between the first light direction and the second light direction.

[0087] The electronic device 10 can perform reparameterization processing on an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path to obtain a reparameterized integral function.

[0088] The electronic device 10 can construct the target three-dimensional model according to the reparameterization integral function corresponding to the plurality of light ray paths.

[0089] It should be noted that, in the case of the electronic device 10 being a terminal device, the electronic device 10 can perform display processing on the target three-dimensional model. As shown in Figure 1 In the case of the electronic device 10 being a server, the electronic device 10 can send the target three-dimensional model to the terminal device 20, so that the terminal device 20 performs display processing on the target three-dimensional model.

[0090] It should be noted that, Figure 1 This is only a scenario diagram of a three-dimensional model construction method provided by the embodiments of the present application, and the embodiments of the present application do not limit the actual form of various components included in the Figure 1 nor limit the interaction mode between the components in the Figure 1 application, which can be set according to actual needs in the application of the scheme.

[0091] The technical scheme of the present application will be described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described in detail in some embodiments.

[0092] Figure 2a A flow diagram of a three-dimensional model construction method provided by an embodiment of the present application. Referring to Figure 2a , the method specifically includes the following steps:

[0093] S201: Obtain a light ray path and split the light ray path to obtain at least one sub-path.

[0094] In this embodiment, the electronic device can determine a light ray path corresponding to a target pixel point. Figure 2b A schematic diagram of a light ray path provided by an embodiment of the present application. As shown in Figure 2b , a light ray path corresponding to a target pixel point is PE→PD→PC→PB→PA(x0, y0).

[0095] The electronic device can perform splitting processing on the light ray path to obtain at least one sub-path.

[0096] In one implementation, the electronic device can obtain the path length of the light ray path, and perform splitting processing on the light ray path according to the path length of the light ray path to obtain at least one sub-path.

[0097] For example, as shown in Figure 2bAs shown, the electronic device can split the light path according to the path length of 4, and obtain 4 sub-paths, namely sub-path 1 (the path between PA (x0, y0) and PB), sub-path 2 (the path between PB and PC), sub-path 3 (the path between PC and PD), and sub-path 4 (the path between PD and PE).

[0098] S202: For any subpath, perform differentiable mapping on any second light direction in the second integration domain, and determine a first light direction corresponding to the second light direction from multiple light directions in the first integration domain.

[0099] In this embodiment, for any sub-path, the sub-path corresponds to a shading point. It can be understood that the shading point is the end point of the light transmission of the sub-path.

[0100] The electronic device may determine a first integration range of the shading point and a second integration range of the shading point.

[0101] It should be noted that the first integration domain is the effective integration domain of the shading point when the scene parameters are updated; the second integration domain is the effective integration domain of the shading point after the scene parameters are updated.

[0102] It should be noted that the first integration domain is the integration domain in the target integration domain before the scene parameters are updated, and the irradiance of the shading point is not 0. The second integration domain is the integration domain in the target integration domain before the scene parameters are updated, and the irradiance of the shading point is not 0.

[0103] It should be noted that the target integration domain is the integration domain of the shading point (also referred to as the ray direction integration domain). In one implementation, the target integration domain can be a spherical integration domain. In another implementation, the target integration domain can be a planar integration domain. It should be noted that when the shading point is a pixel, the integration domain of the shading point, namely the target integration domain, can be a planar integration domain.

[0104] Next, a process of determining the first integration domain and the second integration domain by the electronic device is described.

[0105] In one implementation:

[0106] Figure 2c A schematic diagram of a scene of a single triangle unit provided in an embodiment of the present application.

[0107] like Figure 2c As shown, in the case where the target three-dimensional object before the scene parameter update is a single triangle unit, the target three-dimensional object before the scene parameter update is a single triangle unit:

[0108] The electronic device can project the target three-dimensional object before the scene parameter is updated to the target integral domain to obtain a first integral domain.

[0109] The electronic device can project the target three-dimensional object after the scene parameter is updated to the target integral domain to obtain a second integral domain.

[0110] In an implementation manner:

[0111] Figure 2d A scene schematic diagram of a double-triangle unit is provided for an embodiment of the present application.

[0112] In a case where the target three-dimensional object before the scene parameter is updated is two triangle units (including a shielding triangle unit and a shielded triangle unit), and the target three-dimensional object after the scene parameter is updated is two triangle units (including a shielding triangle unit and a shielded triangle unit):

[0113] The electronic device can project the target three-dimensional object before the scene parameter is updated to the target integral domain to obtain a first initial integral domain.

[0114] The electronic device can perform segmentation processing on the first initial integral domain to obtain a plurality of segmented integral domains. In an implementation manner, the electronic device can determine a first quantity of vertices of the shielding triangle unit that are inside the shielded triangle unit, a second quantity of vertices of the shielded triangle unit that are outside the shielding triangle, and a third quantity of edge intersection points of the two triangles. The electronic device can determine a segmentation manner according to the first quantity, the second quantity, and the third quantity by querying a preset relationship (the preset relationship indicates a corresponding relationship between the first quantity, the second quantity, and the third quantity and the segmentation manner). The electronic device can perform segmentation processing on the first initial integral domain according to the segmentation manner to obtain a plurality of segmented integral domains. Exemplarily, Figure 2e A schematic diagram of a plurality of segmentation manners is provided for an embodiment of the present application.

[0115] The electronic device can determine one of the plurality of segmented integral domains as the first integral domain.

[0116] The electronic device can project the target three-dimensional object after the scene parameter is updated to the target integral domain to obtain a second initial integral domain.

[0117] The electronic device can perform segmentation processing on the second initial integral domain to obtain a plurality of segmented integral domains.

[0118] The electronic device can determine one of the plurality of segmented integral domains, which corresponds to the first integral domain, as the second integral domain. Exemplarily, Figure 2f A schematic diagram of a corresponding relationship of a segmented integral domain is provided for an embodiment of the present application. As Figure 2fAs shown, the first initial integral domain corresponds to the second initial integral domain; the segmentation manner of the first initial integral domain corresponds to the segmentation manner of the second initial integral domain; and each segmented integral domain obtained by segmenting the first initial integral domain corresponds to each segmented integral domain obtained by segmenting the second initial integral domain.

[0119] In an implementation manner, the electronic device can perform the following operations.

[0120] Figure 2g A scene schematic diagram of the plurality of triangular units is provided for the embodiments of the present application.

[0121] In a case where the target three-dimensional object before the scene parameter is updated is a plurality of triangular units and the target three-dimensional object after the scene parameter is updated is a plurality of triangular units, the electronic device can project the target three-dimensional object before the scene parameter is updated to the target integral domain to obtain a first initial integral domain.

[0122] The electronic device can project the target three-dimensional object after the scene parameter is updated to the target integral domain to obtain a second initial integral domain.

[0123] The electronic device can project the target three-dimensional object after the scene parameter is updated to the target integral domain to obtain a second initial integral domain.

[0124] The electronic device can perform segmentation processing on the first initial integral domain to obtain a plurality of segmented integral domains. In an implementation manner, the electronic device can determine a segmentation manner based on the k-d data structure, and perform segmentation processing on the first initial integral domain according to the segmentation manner to obtain the plurality of segmented integral domains.

[0125] In addition, the electronic device can determine one of the plurality of segmented integral domains as the first integral domain.

[0126] The electronic device can project the target three-dimensional object after the scene parameter is updated to the target integral domain to obtain a second initial integral domain.

[0127] The electronic device can perform segmentation processing on the second initial integral domain to obtain a plurality of segmented integral domains.

[0128] The electronic device can determine one of the plurality of segmented integral domains, which corresponds to the first integral domain, as the second integral domain.

[0129] For any sub-path, the electronic device can perform differentiable mapping on any second ray direction in the second integral domain to determine a first ray direction corresponding to the second ray direction from a plurality of ray directions in the first integral domain.

[0130] Next, the process that the electronic device performs differentiable mapping on any second ray direction in the second integral domain to determine a first ray direction corresponding to the second ray direction from a plurality of ray directions in the first integral domain is described.

[0131] In an implementation manner:

[0132] The electronic device can determine a first segmentation unit corresponding to the sub-path and a second segmentation unit. The first segmentation unit belongs to the target three-dimensional object before the scene parameter is updated, and the second segmentation unit belongs to the target three-dimensional object after the scene parameter is updated. It should be noted that the target three-dimensional object corresponds to the sub-path. It should also be noted that the first segmentation unit corresponds to the second segmentation unit, in other words, the first segmentation unit deforms into the second segmentation unit in the case of updating the scene parameter.

[0133] The electronic device can determine the position information of the second mapping point of the second light direction mapping to the second segmentation unit.

[0134] The electronic device can determine the position information of the second centroid point of the second segmentation unit, and determine the position information of the first centroid point of the first segmentation unit.

[0135] The electronic device can determine the position information of the first mapping point according to the position information of the second mapping point, the position information of the second centroid point, and the position information of the first centroid point.

[0136] The electronic device can determine the first light direction from the plurality of light directions included in the first integral domain according to the position information of the first mapping point. The first mapping point is a mapping point of the first light direction mapping to the first segmentation unit.

[0137] S203: Construct a differential homeomorphism corresponding to the sub-path.

[0138] In this embodiment, the electronic device can construct a differential homeomorphism corresponding to the sub-path after obtaining the first light direction and the second light direction.

[0139] The differential homeomorphism indicates the corresponding relationship between the first light direction and the second light direction.

[0140] For example, in the case that the light path is split into n sub-paths, the differential homeomorphism corresponding to the first sub-path is E(1), the differential homeomorphism corresponding to the second sub-path is E(2), and the differential homeomorphism corresponding to the n-th sub-path is E(n).

[0141] E(1) is (x0(π), y0(π)) = E(1)(x0(π0), y0(π0)).

[0142] E(2) is w(π) = E(2)(x0(π0), y0(π0), w(π0), π).

[0143] E(n) is w n-1 (π) = E(n)(w n-2 (π), w n-1 (π0), π).

[0144] It should be noted that π0 is the scene parameter before updating; π is the scene parameter after updating, and w is the light direction.

[0145] S204: According to the differential homeomorphism corresponding to at least one sub-path, the integral function corresponding to the light path is re-parameterized to obtain a re-parameterized integral function.

[0146] In this embodiment, the electronic device can perform re-parameterization processing on the integral function corresponding to the light path according to the differential homeomorphism corresponding to at least one sub-path, to obtain a re-parameterized integral function.

[0147] In an implementation manner:

[0148] For any sub-path, the electronic device can construct the Jacobian corresponding to the sub-path.

[0149] The electronic device can perform re-parameterization processing on the integral function corresponding to the light path according to the differential homeomorphism corresponding to at least one sub-path and the Jacobian corresponding to at least one sub-path, to obtain a re-parameterized integral function corresponding to the light path.

[0150] It can be understood that the electronic device can perform composite processing on the differential homeomorphism corresponding to at least one sub-path to obtain a target differential homeomorphism.

[0151] Exemplarily, in the case that the light path is split into n sub-paths, the differential homeomorphism corresponding to the sub-path 1 is E(1), the differential homeomorphism corresponding to the sub-path 2 is E(2), …, and the differential homeomorphism corresponding to the sub-path n is E(n).

[0152] The electronic device can perform composite processing on E(1), E(2), …, and E(n) to obtain a target differential homeomorphism.

[0153] The target differential homeomorphism is:

[0154] (x0(π), y0(π), w1(π), …, w n-1 (π))

[0155] = (E(1)(x0(π0), y0(π0)), E(2)(x0(π0), y0(π0), w(π0), π),

[0156] …, E(n)(w n-2 (π), w n-1 (π0), π))

[0157] The electronic device can perform composite processing on the Jacobian corresponding to at least one sub-path to obtain a target Jacobian.

[0158] The electronic device can perform reparameterization processing on the integral function corresponding to the light path according to the target differential diffeomorphism and the target Jacobian, to obtain a reparameterized integral function corresponding to the light path.

[0159] S205: Construct the target three-dimensional model according to the reparameterized integral function.

[0160] In this embodiment, the electronic device can construct the target three-dimensional model according to the reparameterized integral function.

[0161] In an implementation manner:

[0162] The target pixel point corresponds to a plurality of light paths.

[0163] The electronic device can determine the target reparameterized integral function according to the reparameterized integral functions corresponding to the plurality of light paths after obtaining the reparameterized integral functions corresponding to the plurality of light paths.

[0164] The target reparameterized integral function is used to calculate the irradiance of the target pixel point, and the reparameterized integral functions corresponding to the light paths are used to calculate the contribution values of the light paths to the irradiance of the target pixel point.

[0165] The electronic device can perform differential processing on the target reparameterized integral function to obtain an i-th first gradient value.

[0166] The electronic device can obtain an i-th scene parameter, and determine an i-th two-dimensional image according to the target integral function (the target integral function is determined according to the integral functions corresponding to the plurality of light paths) and the i-th scene parameter.

[0167] The electronic device can determine an i-th second gradient value according to the i-th two-dimensional image and the original two-dimensional image.

[0168] The electronic device can obtain an i-th scene parameter gradient value according to the i-th first gradient value and the i-th second gradient value.

[0169] The electronic device can determine an (i+1)-th scene parameter according to the i-th scene parameter gradient value.

[0170] The electronic device can determine an (i+1)-th two-dimensional image according to the target integral function and the (i+1)-th scene parameter.

[0171] The electronic device can determine an (i+1)-th second gradient value according to the (i+1)-th two-dimensional image and the original two-dimensional image.

[0172] The electronic device can determine the (i+1)-th scene parameter as a target scene parameter in a case where the (i+1)-th second gradient value satisfies a preset condition, and construct a target three-dimensional model corresponding to the original two-dimensional image according to the target scene parameter. Alternatively,

[0173] When determining that the (i+1)th second gradient value does not satisfy a preset condition, the electronic device may obtain the (i+1)th scene parameter gradient value according to the obtained (i+1)th first gradient value and (i+1)th second gradient value.

[0174] It should be noted that i is a positive integer greater than 0.

[0175] Beneficial effects of this embodiment: In this embodiment, an electronic device can obtain a light path and split the light path to obtain at least one subpath. For any subpath, the electronic device can perform a differentiable mapping on any second light direction in a second integral domain, and determine the first light direction corresponding to the second light direction from multiple light directions in the first integral domain. The first integral domain is the integral domain of the shading points corresponding to the subpath before the scene parameters are updated; the second integral domain is the integral domain of the shading points after the scene parameters are updated. The electronic device can construct a diffeomorphism corresponding to the subpath; the diffeomorphism indicates the correspondence between the first light direction and the second light direction. The electronic device can reparameterize the integral function corresponding to the light path based on the diffeomorphism corresponding to at least one subpath to obtain a reparameterized integral function. The electronic device can construct a target three-dimensional model based on the reparameterized integral function. This method improves the reparameterization rate of the integral function, thereby improving the efficiency of constructing the target three-dimensional model. In addition, this method reduces the variance of the scene parameter gradient information obtained based on the reparameterized integral function.

[0176] The following describes the process of constructing a diffeomorphism corresponding to a subpath of an electronic device through a second method embodiment.

[0177] Figure 3 This is a flow chart of a second embodiment of a method for constructing a three-dimensional model provided in an embodiment of the present application. Figure 3 , the method specifically comprises the following steps:

[0178] S301: Determine a first segmentation unit and a second segmentation unit corresponding to a subpath.

[0179] In this embodiment, the electronic device may determine a first segmentation unit and a second segmentation unit corresponding to the subpath. In one implementation, the first segmentation unit may be a triangle unit, and the second segmentation unit may be a triangle unit.

[0180] It should be noted that the first segmentation unit represents the target 3D object before the scene parameters are updated, while the second segmentation unit represents the target 3D object after the scene parameters are updated. It should also be noted that the target 3D object corresponds to a subpath. In other words, the target 3D object is the starting point for the light transmission of the subpath.

[0181] It should be further noted that the first segmentation unit is a segmentation unit corresponding to the first integral domain among a plurality of segmentation units included in the target three-dimensional object before the scene parameter is updated. The second segmentation unit is a segmentation unit corresponding to the second integral domain among a plurality of segmentation units included in the target three-dimensional object after the scene parameter is updated.

[0182] S302: Determine the position information of the second mapping point of the second light direction mapping to the second segmentation unit.

[0183] In this embodiment, the electronic device can determine the position information of the second mapping point of the second light direction mapping to the second segmentation unit.

[0184] S303: Determine the position information of the second centroid point of the second segmentation unit, and determine the position information of the first centroid point of the first segmentation unit.

[0185] In this embodiment, for the second segmentation unit, the electronic device can obtain the position information of the second segmentation unit after obtaining the second segmentation unit. Exemplarily, the position information of the second segmentation unit is (p0(π), p1(π), p2(π)). Wherein, π is the updated scene parameter.

[0186] The electronic device can determine the position information of the second centroid point according to the position information of the second segmentation unit. Wherein, the second centroid point is the centroid point of the second segmentation unit.

[0187] The electronic device can obtain the position information of the first segmentation unit after obtaining the first segmentation unit. Exemplarily, the position information of the first segmentation unit is (p0(π0), p1(π0), p2(π0)). Wherein, π0 is the scene parameter before the update.

[0188] The electronic device can determine the position information of the first centroid point according to the position information of the first segmentation unit. Wherein, the first centroid point is the centroid point of the first segmentation unit.

[0189] S304: Determine the position information of the first mapping point according to the position information of the second mapping point, the position information of the second centroid point and the position information of the first centroid point.

[0190] In this embodiment, the electronic device can determine the position information of the first mapping point according to the position information of the second mapping point, the position information of the second centroid point and the position information of the first centroid point.

[0191] In one implementation manner:

[0192] The electronic device can determine the relative position according to the position information of the second mapping point and the position information of the second centroid point.

[0193] The electronic device can determine the position information of the first mapping point according to the relative position and the position information of the first centroid point.

[0194] S305: determining the first light direction from the plurality of light directions included in the first integral domain according to the position information of the first mapping point.

[0195] In this embodiment, the electronic device can determine the first light direction from the plurality of light directions included in the first integral domain according to the position information of the first mapping point.

[0196] It can be understood that the electronic device can determine the first light direction according to the position information of the first mapping point in a reverse light tracing manner.

[0197] It should be noted that the first mapping point is a mapping point of the first light direction to the first partition unit.

[0198] Advantages of the present embodiment: In this embodiment, the electronic device can determine the first partition unit and the second partition unit corresponding to the sub-path; the first partition unit corresponds to the target three-dimensional object before the scene parameter is updated, the second partition unit corresponds to the target three-dimensional object after the scene parameter is updated, and the target three-dimensional object corresponds to the sub-path. The electronic device can determine the position information of the second mapping point of the second light direction to the second partition unit. The electronic device can determine the position information of the second centroid point of the second partition unit and the position information of the first centroid point of the first partition unit. The electronic device can determine the position information of the first mapping point according to the position information of the second mapping point, the position information of the second centroid point and the position information of the first centroid point. The electronic device can determine the first light direction from the plurality of light directions included in the first integral domain according to the position information of the first mapping point; the first mapping point is a mapping point of the first light direction to the first partition unit. Through the above manner, the corresponding relationship between the first light direction and the second light direction can be accurately established, so that the integral domain of the integral function based on the differential homeomorphism reparameterization is switched from the second integral domain (the effective integral domain of the shading point after the scene parameter is updated) to the first integral domain (the effective integral domain of the shading point before the scene parameter is updated), so that the discontinuous boundary of the integral function after reparameterization will no longer change with the change of the scene parameter, and the construction accuracy of the three-dimensional geometric model is improved.

[0199] Figure 4 A flowchart of a three-dimensional model construction method embodiment provided by the present application is shown in FIG. 3. Figure 4 The method specifically includes the following steps:

[0200] S401: obtaining a light path and splitting the light path to obtain at least one sub-path.

[0201] In this embodiment, the electronic device can obtain a light path and split the light path to obtain at least one sub-path.

[0202] The specific implementation process is the same as that of S201, which will not be described here.

[0203] S402: For any sub-path, differentiable mapping is performed on any second light direction in the second integral domain, and a first light direction is determined from a plurality of light directions in the first integral domain.

[0204] In this embodiment, for any sub-path, the electronic device can perform differentiable mapping on any second light direction in the second integral domain, and determine a first light direction from a plurality of light directions in the first integral domain.

[0205] The first integral domain is the effective integral domain of the shading point before the scene parameter is updated; the second integral domain is the effective integral domain of the shading point after the scene parameter is updated; and the shading point corresponds to the sub-path.

[0206] The specific implementation process is the same as that of S202, which will not be described here.

[0207] S403: Constructing a differential homeomorphism corresponding to the sub-path.

[0208] In this embodiment, the electronic device can construct a differential homeomorphism corresponding to the sub-path.

[0209] The differential homeomorphism indicates the corresponding relationship between the first light direction and the second light direction.

[0210] The specific implementation process is the same as that of S203, which will not be described here.

[0211] S404: Constructing a Jacobian item corresponding to the sub-path according to the first segmentation unit and the second segmentation unit.

[0212] In this embodiment, for any sub-path, the electronic device can construct a Jacobian item corresponding to the sub-path according to the first segmentation unit and the second segmentation unit.

[0213] Next, the process of constructing a Jacobian item corresponding to the sub-path according to the first segmentation unit and the second segmentation unit by the electronic device will be described.

[0214] In one implementation manner:

[0215] The electronic device can determine whether the first integral domain and the second integral domain belong to a planar integral domain.

[0216] In the case where the first integral domain and the second integral domain do not belong to a planar integral domain:

[0217] The electronic device can determine the Jacobian corresponding to the sub-path according to the acquired position information of the first segmentation unit, the acquired position information of the second segmentation unit, the position information of the first mapping point, and the position information of the second mapping point.

[0218] Exemplarily, the electronic device can calculate the Jacobian based on the following formula:

[0219]

[0220] It should be noted that the position information of the first segmentation unit is (p0(π0), p1(π0), p2(π0)); the position information of the second segmentation unit is (p0(π), p1(π), p2(π)); p(π) is used to indicate the position information of the second mapping point; p(π0) is used to indicate the position information of the first mapping point.

[0221] In the case where the first integral domain and the second integral domain belong to a planar integral domain:

[0222] The electronic device can project the first segmentation unit onto a preset image plane to obtain a first projection unit;

[0223] The electronic device can project the second segmentation unit onto a preset image plane to obtain a second projection unit;

[0224] The electronic device can determine the Jacobian corresponding to the sub-path according to the acquired position information of the first projection unit, and the acquired position information of the second projection unit.

[0225] Exemplarily, the electronic device can calculate the Jacobian based on the following formula:

[0226]

[0227] It should be noted that (p 0s (π), p 1s (π), p 2s (π)) indicates the position information of the second projection unit; (p 0s (π0), p 1s (π0), p 2s (π0)) indicates the position information of the first projection unit.

[0228] S405: According to the differential homeomorphism corresponding to at least one sub-path and the Jacobian corresponding to at least one sub-path, the integral function corresponding to the light path is re-parameterized to obtain a re-parameterized integral function corresponding to the light path.

[0229] In the embodiment, the electronic device can perform reparameterization processing on the integral function corresponding to the light path according to the at least one differential homeomorphism corresponding to the at least one sub-path and the at least one Jacobian corresponding to the at least one sub-path, to obtain a reparameterization integral function corresponding to the light path.

[0230] In an implementation manner, the electronic device can perform reparameterization processing on the integral function corresponding to the light path according to the at least one differential homeomorphism corresponding to the at least one sub-path and the at least one Jacobian corresponding to the at least one sub-path, to obtain a reparameterization integral function corresponding to the light path.

[0231] The electronic device can perform composition processing on the at least one differential homeomorphism corresponding to the at least one sub-path to obtain a target differential homeomorphism.

[0232] The electronic device can perform composition processing on the at least one Jacobian corresponding to the at least one sub-path to obtain a target Jacobian.

[0233] The electronic device can perform reparameterization processing on the integral function corresponding to the light path according to the target differential homeomorphism and the target Jacobian, to obtain a reparameterization integral function corresponding to the light path.

[0234] S406: constructing a target three-dimensional model according to the reparameterization integral function.

[0235] In the embodiment, the electronic device can construct a target three-dimensional model according to the reparameterization integral function.

[0236] The specific implementation process is the same as that of S205, which will not be described here.

[0237] In the embodiment, the electronic device can obtain a light path and perform splitting processing on the light path to obtain at least one sub-path. For any sub-path, the electronic device can perform differentiable mapping on any second light direction in a second integral domain to determine a first light direction from a plurality of light directions in a first integral domain; the first integral domain is an integral domain of a shading point before scene parameter update; the second integral domain is an integral domain of the shading point after scene parameter update; the shading point corresponds to the sub-path. The electronic device can construct a differential homeomorphism corresponding to the sub-path; the differential homeomorphism indicates the correspondence between the first light direction and the second light direction. The electronic device can construct a Jacobian corresponding to the sub-path according to the first segmentation unit and the second segmentation unit. The electronic device can perform reparameterization processing on the integral function corresponding to the light path according to the at least one differential homeomorphism corresponding to the at least one sub-path and the at least one Jacobian corresponding to the at least one sub-path, to obtain a reparameterization integral function corresponding to the light path. The electronic device can construct a target three-dimensional model according to the reparameterization integral function. In the above manner, the reparameterization rate and accuracy of the integral function are improved, and the construction efficiency and accuracy of the target three-dimensional model are improved.

[0238] Figure 5 A structural schematic diagram of a three-dimensional model construction device provided by an embodiment of the present application is shown in FIG. 1. Figure 5As shown, the three-dimensional model construction device 50 can include a processing module 51 and a construction module 52.

[0239] The processing module 51 is configured to obtain a light path and split the light path to obtain at least one sub-path.

[0240] The processing module 51 is further configured to, for any sub-path, perform differentiable mapping on any second light direction in a second integral domain to determine a first light direction corresponding to the second light direction from a plurality of light directions in a first integral domain; the first integral domain is an effective integral domain of a shading point corresponding to the sub-path before scene parameter update; and the second integral domain is an effective integral domain of the shading point after scene parameter update.

[0241] The processing module 51 is further configured to construct a differential homeomorphism corresponding to the sub-path; the differential homeomorphism indicates a corresponding relationship between the first light direction and the second light direction.

[0242] The processing module 51 is further configured to perform reparameterization processing on an integral function corresponding to the light path according to the differential homeomorphism corresponding to the at least one sub-path to obtain a reparameterized integral function.

[0243] The construction module 52 is configured to construct a target three-dimensional model according to the reparameterized integral function.

[0244] The three-dimensional model construction device provided in the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and has similar implementation principles and beneficial effects, which will not be described in detail here.

[0245] In an implementation manner, the processing module 51 is specifically configured to include the following steps.

[0246] Determine a first segmentation unit and a second segmentation unit corresponding to the sub-path; the first segmentation unit belongs to a target three-dimensional object before scene parameter update, and the second segmentation unit belongs to a target three-dimensional object after scene parameter update; and the target three-dimensional object corresponds to the sub-path.

[0247] Determine position information of a second mapping point of the second light direction in the second segmentation unit;

[0248] Determine position information of a second centroid point of the second segmentation unit and position information of a first centroid point of the first segmentation unit;

[0249] Determine position information of a first mapping point according to the position information of the second mapping point, the position information of the second centroid point, and the position information of the first centroid point;

[0250] Determine the first light direction from a plurality of light directions included in the first integral domain according to the position information of the first mapping point; and the first mapping point is a mapping point of the first light direction in the first segmentation unit.

[0251] The three-dimensional model construction device provided by the embodiments of the present application can execute the technical solutions shown in the method embodiments, and the implementation principles and beneficial effects are similar, and will not be repeated here.

[0252] In an implementation manner, the processing module 51 is specifically configured to:

[0253] According to the position information of the second mapping point and the position information of the second centroid point, the relative position is determined.

[0254] According to the relative position and the position information of the first centroid point, the position information of the first mapping point is determined.

[0255] The three-dimensional model construction device provided by the embodiments of the present application can execute the technical solutions shown in the method embodiments, and the implementation principles and beneficial effects are similar, and will not be repeated here.

[0256] In an implementation manner, the processing module 51 is specifically configured to:

[0257] According to the first segmentation unit and the second segmentation unit, the Jacobian term corresponding to the sub-path is constructed.

[0258] According to the differential homeomorphism corresponding to at least one sub-path and the Jacobian term corresponding to at least one sub-path, the integral function corresponding to the light path is re-parameterized to obtain the re-parameterized integral function corresponding to the light path.

[0259] The three-dimensional model construction device provided by the embodiments of the present application can execute the technical solutions shown in the method embodiments, and the implementation principles and beneficial effects are similar, and will not be repeated here.

[0260] In an implementation manner, the processing module 51 is specifically configured to:

[0261] Determine whether the first integral domain and the second integral domain belong to a planar integral domain.

[0262] In the case that the first integral domain and the second integral domain do not belong to the planar integral domain, according to the obtained position information of the first segmentation unit, the obtained position information of the second segmentation unit, the position information of the first mapping point and the position information of the second mapping point, the Jacobian term corresponding to the sub-path is determined.

[0263] The three-dimensional model construction device provided by the embodiments of the present application can execute the technical solutions shown in the method embodiments, and the implementation principles and beneficial effects are similar, and will not be repeated here.

[0264] In an implementation manner, the processing module 51 is further configured to:

[0265] In a case where the first integral domain and the second integral domain belong to a planar integral domain, the first partition unit is projected onto a preset image plane to obtain a first projection unit;

[0266] The second partition unit is projected onto the preset image plane to obtain a second projection unit;

[0267] According to the position information of the first projection unit and the position information of the second projection unit, a Jacobian item corresponding to the sub-path is determined.

[0268] The three-dimensional model construction device provided in the embodiments of the present application can execute the technical solutions shown in the above method embodiments, and has similar implementation principles and beneficial effects, which will not be described here in detail.

[0269] Figure 6 A structural diagram of an electronic device is provided in the present application. As shown in Figure 6 The electronic device 60 includes a processor 61 and a memory 62. The processor 61 is in communication connection with the memory 62, and the memory 62 is configured to store computer execution instructions; the processor 61 is configured to execute the technical solutions in any of the above method embodiments by executing the computer execution instructions stored in the memory 62.

[0270] Optionally, the memory 62 can be independent or integrated with the processor 61. Optionally, when the memory 62 is independent of the processor 61, the electronic device 60 can further include a bus 63 for connecting the above devices.

[0271] The electronic device is configured to execute the technical solutions in any of the above method embodiments, and has similar implementation principles and technical effects, which will not be described here in detail.

[0272] The embodiments of the present application further provide a computer readable storage medium, and the computer readable storage medium stores computer execution instructions. When the computer execution instructions are executed by a processor, the computer execution instructions are configured to implement the technical solutions provided in any of the above method embodiments.

[0273] The embodiments of the present application further provide a computer program product, which includes a computer program. When the computer program is executed by a processor, the computer program is configured to implement the technical solutions provided in the above method embodiments.

[0274] It should be noted that, for the above method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the action order described, because according to the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0275] It should be understood that the steps in the flowchart diagrams are not necessarily executed in the order shown in the flowcharts. Unless otherwise specified, the steps of the flowcharts are not necessarily executed in the order shown in the flowcharts. The steps of the flowcharts can be executed in other orders. Moreover, at least some of the steps in the flowcharts can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution of the sub-steps or stages can not necessarily be sequential, but can be performed alternately or alternately with at least part of other steps or sub-steps or stages of other steps.

[0276] It should be understood that the above-described device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-described embodiments is only a logical functional division, and actual implementation can have another division manner. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be omitted or not executed.

[0277] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application can be integrated in one unit / module, or each unit / module can exist physically, or two or more units / modules can be integrated together. The integrated unit / module can be realized in the form of hardware or in the form of a software program module.

[0278] If the integrated unit / module is realized in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the processor can be any appropriate hardware processor, such as CPU, GPU, FPGA, DSP and ASIC, etc. Unless otherwise specified, the storage unit can be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory RRAM (Resistive Random Access Memory), dynamic random access memory DRAM (Dynamic Random Access Memory), static random access memory SRAM (Static Random-Access Memory), enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), high bandwidth memory HBM (High-Bandwidth Memory), hybrid memory cube HMC (Hybrid Memory Cube), etc.

[0279] If the integrated units / modules are implemented in the form of software program modules and sold or used as independent products, they can be stored in a computer readable memory. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a memory and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned memory includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0280] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application

[0281] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. The application is intended to cover any variations, uses or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice in the art to which the application pertains or can relate. The specification and examples are to be regarded as exemplary only, and the true scope and spirit of the application are indicated by the following claims.

[0282] It should be understood that the present application is not limited to the precise construction that has been described above and illustrated in the accompanying drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the application is limited only by the claims that follow.

Claims

1. A method for constructing a three-dimensional model, characterized in that: include: Obtaining a light path, and splitting the light path to obtain at least one sub-path; For any subpath, perform differentiable mapping on any second ray direction in the second integration domain, and determine the first ray direction corresponding to the second ray direction from multiple ray directions in the first integration domain; the first integration domain is the effective integration domain of the shading point corresponding to the subpath before the scene parameters are updated; the second integration domain is the effective integration domain of the shading point after the scene parameters are updated; Constructing a diffeomorphism corresponding to the subpath; the diffeomorphism indicates a correspondence between the first light direction and the second light direction; reparameterizing the integral function corresponding to the light path according to a diffeomorphism corresponding to at least one subpath to obtain a reparameterized integral function; A target three-dimensional model is constructed according to the reparameterized integral function.

2. The method according to claim 1, characterized in that The performing differentiable mapping on any second light direction in the second integral domain and determining, from a plurality of light directions in the first integral domain, a first light direction corresponding to the second light direction includes: Determine a first segmentation unit and a second segmentation unit corresponding to the subpath; the first segmentation unit belongs to the target three-dimensional object before the scene parameters are updated, the second segmentation unit belongs to the target three-dimensional object after the scene parameters are updated, and the target three-dimensional object corresponds to the subpath; Determine position information of a second mapping point of the second segmentation unit mapped to the second light direction; Determining position information of a second centroid of the second segmentation unit, and determining position information of a first centroid of the first segmentation unit; determining the position information of the first mapping point based on the position information of the second mapping point, the position information of the second centroid point, and the position information of the first centroid point; The first light direction is determined from a plurality of light directions included in the first integral domain according to the position information of the first mapping point; the first mapping point is a mapping point where the first light direction is mapped to the first segmentation unit.

3. The method according to claim 2, characterized in that The determining the position information of the first mapping point according to the position information of the second mapping point, the position information of the second centroid point, and the position information of the first centroid point includes: determining a relative position based on the position information of the second mapping point and the position information of the second centroid; The position information of the first mapping point is determined according to the relative position and the position information of the first centroid point.

4. The method according to claim 2, characterized in that The step of reparameterizing the integral function corresponding to the light path according to the diffeomorphism corresponding to at least one subpath to obtain the reparameterized integral function corresponding to the light path includes: Constructing a Jacobian term corresponding to the subpath according to the first segmentation unit and the second segmentation unit; According to the differential homeomorphism corresponding to at least one subpath and the Jacobian term corresponding to at least one subpath, the integral function corresponding to the light path is reparameterized to obtain the reparameterized integral function corresponding to the light path.

5. The method according to claim 4, characterized in that The constructing the Jacobian term corresponding to the subpath according to the first segmentation unit and the second segmentation unit includes: determining whether the first integration domain and the second integration domain belong to a plane integration domain; When the first integral domain and the second integral domain do not belong to a planar integral domain, the Jacobian term corresponding to the subpath is determined according to the acquired position information of the first segmentation unit, the acquired position information of the second segmentation unit, the position information of the first mapping point, and the position information of the second mapping point.

6. The method according to claim 5, characterized in that The method further comprises: When the first integral domain and the second integral domain belong to a plane integral domain, projecting the first segmentation unit onto a preset image plane to obtain a first projection unit; Projecting the second segmentation unit onto the preset image plane to obtain a second projection unit; The Jacobian term corresponding to the subpath is determined according to the acquired position information of the first projection unit and the acquired position information of the second projection unit.

7. A three-dimensional model construction device, characterized in that: include: a processing module, configured to obtain a light path and split the light path to obtain at least one sub-path; The processing module is further configured to perform differentiable mapping on any second ray direction in a second integration domain for any subpath, and determine a first ray direction corresponding to the second ray direction from a plurality of ray directions in a first integration domain; the first integration domain being a valid integration domain of a shading point corresponding to the subpath before scene parameters are updated; and the second integration domain being a valid integration domain of the shading point after scene parameters are updated; The processing module is further configured to construct a diffeomorphism corresponding to the subpath; the diffeomorphism indicates a correspondence between the first light direction and the second light direction; The processing module is further configured to perform a reparameterization process on the integral function corresponding to the light path according to a diffeomorphism corresponding to at least one sub-path, to obtain a reparameterized integral function; A construction module is used to construct a target three-dimensional model according to the reparameterized integral function.

8. An electronic device, characterized in that: include: memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The method comprises computer-executable instructions, which, when executed by a processor, enable the method according to any one of claims 1 to 6 to be performed.

Images