Method for converting two-dimensional image into three-dimensional image

Abstract

The embodiment of the invention provides a method for converting a two-dimensional image into a three-dimensional image. In one specific embodiment, the method comprises the following steps: acquiring an image source required by 3D conversion; positioning an object in the current shot picture through the shooting terminal, and obtaining a multi-angle image of at least one to-be-converted object and a positioning result corresponding to each angle image in each to-be-converted object through the shooting terminal; cutting each angle image to obtain a first multi-angle cutting image of each to-be-converted object; and performing 3D conversion on each to-be-converted object to obtain a 3D image. According to the embodiment, the data volume of image processing during three-dimensional conversion is reduced by cutting the image, and meanwhile, when the shooting terminal carries out shooting, the image is already positioned, so that the three-dimensional conversion efficiency can be effectively improved; a user can actually adjust various parameters of the shot image according to the field condition to improve the shooting quality, so that the quality of the obtained three-dimensional image is higher.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing, and more specifically, to a method for converting a two-dimensional image into a three-dimensional image. Background Art

[0002] Converting a two-dimensional image (2D image) to a three-dimensional image (3D image) refers to the process of converting a two-dimensional flat image or view into a three-dimensional model or scene with depth information and stereoscopic effect. This conversion process usually involves multiple steps and complex technologies, aiming to extract enough depth, shape and texture information from two-dimensional pixel data to construct a realistic three-dimensional representation.

[0003] Although significant progress has been made in converting current two-dimensional images into three-dimensional images, conversion efficiency is still a key issue that restricts its widespread application. Summary of the invention

[0004] The purpose of the present disclosure is to provide a method for efficiently converting a two-dimensional image into a three-dimensional image, so as to solve at least one of the problems existing in the prior art.

[0005] In order to achieve the above objectives, the present invention adopts the following technical solutions:

[0006] A first aspect of the present disclosure provides a method for converting a two-dimensional image into a three-dimensional image, comprising:

[0007] Obtain the image source required for 3D conversion through the shooting terminal;

[0008] When the image source is real-time shooting, the object in the current shooting picture is positioned through the shooting terminal, and the multi-angle image of at least one object to be converted and the positioning result corresponding to each angle image of each object to be converted are obtained through the shooting terminal;

[0009] According to the positioning results corresponding to the images at each angle in each object to be converted, the images at each angle in each object to be converted are respectively cropped to obtain first multi-angle cropped images of each object to be converted;

[0010] Based on the first multi-angle cropped images of each object to be converted, 3D conversion is performed on each object to be converted to obtain a 3D image of each object to be converted.

[0011] Furthermore, positioning the object in the current shooting picture by the shooting terminal includes:

[0012] Identify the object at the center of the current shooting picture or identify the shape outline of the object in the user-set area in the current shooting picture in response to the user's setting operation, and use the shape outline of the object as the positioning result of the object in the current shooting picture.

[0013] Furthermore, after positioning the object in the current shooting picture by the shooting terminal, the method further includes:

[0014] Identify the object at the center of the current shooting picture or respond to the user's setting operation to identify the surface material, surface color and surface light of the object in the user-set area in the current shooting picture, and judge and output the type of object in the current shooting picture based on the object's shape outline, surface material, surface color and surface light.

[0015] Furthermore, the step of cropping the images at each angle of the objects to be transformed according to the positioning results corresponding to the images at each angle of the objects to be transformed comprises:

[0016] The images at each angle of each object to be converted are respectively cut according to the shape contours corresponding to the images at each angle of each object to be converted;

[0017] After acquiring, through the shooting terminal, multi-angle images of at least one object to be converted and positioning results corresponding to the images at each angle of each object to be converted, the method further includes:

[0018] Evaluate the quality of each image at each angle of each object to be converted and output the evaluation results.

[0019] Furthermore, when the image source is real-time shooting, the method further includes:

[0020] Acquire and output multiple shooting heights and the maximum relative rotation angle required for two adjacent images shot around the object to be converted at each shooting height through a shooting terminal;

[0021] In response to the user's nth shooting operation, the nth image is acquired, and the relative rotation angle between the nth image shot by the shooting terminal relative to the object to be converted at the current shooting height and the current shooting picture shot by the shooting terminal relative to the object to be converted is less than the maximum relative rotation angle, and the recognition result is output; wherein n is an integer greater than 1;

[0022] The maximum relative rotation angle is greater than 0° and less than or equal to 180°.

[0023] Furthermore, the relative rotation angle between the nth image captured by the shooting terminal relative to the object to be converted at the current shooting height and the current shooting image captured by the shooting terminal relative to the object to be converted includes:

[0024] The shooting terminal is fixed at a corresponding shooting height, and the object to be converted rotates around its longitudinal central axis. The shooting terminal obtains the rotation angle to obtain the nth image taken by the shooting terminal relative to the object to be converted at the current shooting height and the relative rotation angle when the shooting terminal takes the current shooting picture relative to the object to be converted;

[0025] Alternatively, the object to be converted is fixed, the shooting terminal is rotated with the object to be converted as the central axis, and the shooting terminal identifies the nth image captured by the shooting terminal around the object to be converted at the current shooting height and the relative rotation angle of the shooting terminal relative to the object to be converted when the current shooting picture is captured.

[0026] Furthermore, the number of the objects to be converted is multiple;

[0027] After obtaining the first multi-angle cropped image of each object to be converted, the method further includes:

[0028] The first multi-angle cropped images of the objects to be converted are spliced ​​or the first multi-angle cropped images of the objects to be converted are spliced ​​with a preset background image to perform 3D conversion on the spliced ​​objects to be converted to obtain 3D images of the objects to be converted.

[0029] Furthermore, when the image source is a multi-angle pre-stored image of each object to be identified stored in a memory;

[0030] The shape outline of the object to be identified in the pre-stored images at each angle of each object to be identified and located at the center of the current shooting picture is identified and cropped to obtain a second multi-angle cropped image of each object to be identified; based on the second multi-angle cropped image of each object to be converted, each object to be converted is 3D converted to obtain a 3D image of each object to be converted.

[0031] Furthermore, after acquiring a multi-angle image of at least one object to be converted through the shooting terminal, the method further includes:

[0032] Eliminate distortion of multi-angle images of each object to be transformed.

[0033] Furthermore, after obtaining the 3D images of the objects to be converted, the method further includes:

[0034] The 3D image of each object to be converted is output to display the 3D image of each object to be converted.

[0035] The beneficial effects of the present disclosure are as follows:

[0036] When the image source is real-time shooting, the present invention uses the shooting terminal device to identify the object in the shooting picture in real time to locate the object, and crops the image shot by the shooting terminal according to the positioning result; the present invention reduces the amount of data for image processing during three-dimensional conversion by cropping the image, and at the same time, the present invention starts to locate the image when the shooting terminal is shooting, which can effectively improve the efficiency of three-dimensional conversion compared to image processing after obtaining the photo; at the same time, the present invention uses the shooting terminal to shoot the image in real time, and the user can actually adjust the various parameters of the shot image according to the on-site situation and supervise the shooting quality of each photo, so that high-precision and high-image quality photos can be obtained, and the three-dimensional image obtained in this way has higher precision and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The specific implementation methods of the present disclosure are further described in detail below with reference to the accompanying drawings.

[0038] Figure 1 A flowchart of a method for converting a two-dimensional image into a three-dimensional image provided by an embodiment of the present disclosure is shown.

[0039] Figure 2 A schematic structural diagram of a device for a method for converting a two-dimensional image into a three-dimensional image provided by an embodiment of the present disclosure is shown.

[0040] Figure 3 A schematic diagram showing shooting at a certain height when the shooting terminal provided by an embodiment of the present disclosure is a camera.

[0041] Figure 4 A schematic diagram showing the structure of a computer system for implementing the apparatus provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0042] In order to more clearly illustrate the present disclosure, the present disclosure is further described below in conjunction with the embodiments and drawings. Similar components in the drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the content described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present disclosure.

[0043] like Figure 1 and Figure 2 As shown, an embodiment of the present disclosure provides a method for converting a two-dimensional image into a three-dimensional image, comprising:

[0044] Step S1, obtaining an image source required for 3D conversion through a shooting terminal;

[0045] Step S2: when the image source is real-time shooting, positioning the object in the current shooting picture through the shooting terminal and obtaining the multi-angle image of at least one object to be converted and the positioning result corresponding to each angle image of each object to be converted through the shooting terminal;

[0046] Step S3, cropping the images at each angle of each object to be converted according to the positioning results corresponding to the images at each angle of each object to be converted, so as to obtain first multi-angle cropped images of each object to be converted;

[0047] Step S4: performing 3D conversion on each object to be converted based on the first multi-angle cropped image of each object to be converted, to obtain a 3D image of each object to be converted.

[0048] When the image source is real-time shooting, the present invention uses the shooting terminal device to identify the object in the shooting picture in real time to locate the object, and crops the image shot by the shooting terminal according to the positioning result; the present invention reduces the amount of data for image processing during three-dimensional conversion by cropping the image, and at the same time, the present invention starts to locate the image when the shooting terminal is shooting, which can effectively improve the efficiency of three-dimensional conversion compared to image processing after obtaining the photo; at the same time, the present invention uses the shooting terminal to shoot the image in real time, and the user can actually adjust the various parameters of the shot image according to the on-site situation and supervise the shooting quality of each photo, so that high-precision and high-image quality photos can be obtained, and the three-dimensional image obtained in this way has higher precision and quality.

[0049] In this embodiment, step S3 and step S4 are performed by an image conversion module.

[0050] In a possible implementation, positioning the object in the current shooting picture by the shooting terminal includes:

[0051] Identify the object at the center of the current shooting picture or identify the shape outline of the object in the user-set area in the current shooting picture in response to the user's setting operation, and use the shape outline of the object as the positioning result of the object in the current shooting picture.

[0052] After positioning the object in the current shooting picture by the shooting terminal, the method further includes:

[0053] Identify the object at the center of the current shooting picture or respond to the user's setting operation to identify the surface material, surface color and surface light of the object in the user-set area in the current shooting picture, and judge and output the type of object in the current shooting picture based on the object's shape outline, surface material, surface color and surface light.

[0054] The present invention determines the type of an object by identifying the shape outline, surface material, surface color and surface light of the object. The user uses the type as a basis to determine whether the type of image recognition is consistent with the type determined by the user himself. If they are consistent, it means that the image is captured correctly. If they are inconsistent, it will affect the three-dimensional conversion results of subsequent images (for example, the shooting terminal device captures a person, but the user sees an animal. If the user takes a photo at this time, the outline of the image is likely to be inconsistent with the actual outline of the object, which will cause errors in the subsequent image cropping and directly affect the conversion accuracy of the subsequent three-dimensional image). At this time, the user needs to adjust the angle, light, saturation and other parameters of the shooting terminal to make the type of the captured image consistent with the actual photo.

[0055] In this embodiment, the user can identify the object at the center of the shooting screen through the shooting terminal according to specific design requirements, or the user can manually select the object to be converted into a 3D effect, so as to judge and identify the outline of the object to be converted. The present invention can also measure the outline volume of the object and calculate the perspective effect. If accurate data is needed for distance measurement, auxiliary distance measurement tools such as laser distance measurement, distributed distance measurement, etc. can be used. Manual focus or automatic focus can be achieved. The final result is to mark the outline of the photographed object.

[0056] In a possible implementation, the step of cropping the images at each angle of the objects to be transformed according to the positioning results corresponding to the images at each angle of the objects to be transformed includes:

[0057] The images at each angle of each object to be converted are respectively cut according to the shape contours corresponding to the images at each angle of each object to be converted;

[0058] After acquiring, through the shooting terminal, multi-angle images of at least one object to be converted and positioning results corresponding to the images at each angle of each object to be converted, the method further includes:

[0059] Evaluate the quality of each image at each angle of each object to be converted and output the evaluation results.

[0060] In one possible implementation, image quality can be assessed in the form of No Reference (NR) metrics; the quality is evaluated by the processed image itself, without the need for the original image. No Reference metrics include: Natural Image Quality Evaluator (NIQE): Assess quality based on the statistical characteristics of natural images, without the need for any training data or reference images. Visual Information Fidelity (VIF): Simulates the human visual system to assess image quality. Although its calculation is more complex, it can better reflect human visual perception. Blind / No-Reference Image Quality Evaluators (BIQEs): Assess quality by analyzing the characteristics of image content, such as deep learning-based methods, which can automatically learn features related to image quality and predict image quality scores.

[0061] In a possible implementation, when the image source is real-time shooting, the method further includes acquiring and outputting, through a shooting terminal, a plurality of shooting heights and a maximum relative rotation angle required for two adjacent images shot by the shooting terminal relative to the object to be converted at each shooting height;

[0062] In response to the user's nth shooting operation, the nth image is acquired, and the relative rotation angle between the nth image shot by the shooting terminal relative to the object to be converted at the current shooting height and the current shooting picture shot by the shooting terminal relative to the object to be converted is less than the maximum relative rotation angle, and the recognition result is output; wherein n is an integer greater than 1;

[0063] The maximum relative rotation angle is greater than 0° and less than or equal to 180°.

[0064] When taking photos, you can take a single shot or continuous shots after accurate focus. When taking continuous shots, you can just integrate a clear 2D image of that angle.

[0065] Shooting around the object at 360 degrees is the basis for completing the entire 3D image conversion. To achieve 360-degree shooting, at least 2 shots are required, i.e., 180 degrees each, 3 shots are required, 120 degrees each, and 90 degrees each. The more shots are taken, the smaller the angle is, the more the image repetition in adjacent photos is, the more data is obtained, and the better the conversion effect, details, etc. Of course, 2 shots are also acceptable, but the edge data obtained will be very small. The present invention preferably shoots 4 times, each at 90 degrees.

[0066] like Figure 3As shown, in order to achieve better results during 3D conversion, images need to be captured from multiple angles. The most ideal situation is to capture a 360-degree external image of the object, such as shooting at four angles, each 90 degrees, to form a 360-degree surface image of the object.

[0067] When the action or position of the person or object being photographed changes during the photo shooting process, one method is to maintain the original action during the photo shooting and perform 3D conversion. The effect of the 3D image conversion is to convert according to the action of the first image taken. Another method is to convert in such a way that the action of the person or object being photographed will also change when the 3D photo is rotated, that is, to perform 3D conversion according to the actual photo taken. The above methods need to perform reasonable and flexible conversion according to the action of the person or object being photographed, and avoid abrupt jumps.

[0068] In a possible implementation, the relative rotation angle between the nth image captured by the shooting terminal relative to the object to be converted at the current shooting height and the current shooting image captured by the shooting terminal relative to the object to be converted includes:

[0069] The shooting terminal is fixed at a corresponding shooting height, and the object to be converted rotates around its longitudinal central axis. The shooting terminal obtains the rotation angle to obtain the nth image taken by the shooting terminal relative to the object to be converted at the current shooting height and the relative rotation angle when the shooting terminal takes the current shooting picture relative to the object to be converted;

[0070] Alternatively, the object to be converted is fixed, the shooting terminal is rotated with the object to be converted as the central axis, and the shooting terminal identifies the nth image captured by the shooting terminal around the object to be converted at the current shooting height and the relative rotation angle of the shooting terminal relative to the object to be converted when the current shooting picture is captured.

[0071] In the present invention, the shooting terminal can be fixed and the object to be converted can be rotated to shoot, or the object to be converted can be fixed and the shooting terminal can be used to rotate around the object to be converted. Therefore, there are two forms of rotation: one is that the central object and the background rotate together, and the other is that the central object rotates and the background does not rotate. The central object is independently prominent relative to the background, and the central object forms a perspective relationship with the background. The background can be spliced ​​with a 2D image, and can also be converted into a 3D image. The background can also be replaced with other images and converted into a new 3D image together with the central object.

[0072] In the present invention, 3D conversion is performed on each object to be converted based on the first multi-angle cropped image of each object to be converted, and a variety of methods such as a depth map generation algorithm based on machine learning can be used to obtain the 3D image of each object to be converted, which is not limited here.

[0073] The present invention is based on the perspective method or the three-dimensional model construction method to perform 3D conversion:

[0074] The basic definition of perspective: The word "perspective" is derived from the Latin word "perspclre" (see through), which refers to the method or technique of depicting the spatial relationship of objects on a plane or curved surface. The initial study of perspective was to look at the scenery through a transparent plane. Accurately depicting the scenery on this plane will form a perspective view of the scenery. Later, the science of using lines to show the spatial position, outline and projection of objects on a plane according to certain principles was called perspective. Imagine a piece of glass between the painter and the object being painted, fix the position of the eyes (look with one eye), connect the key points of the object with the eyes to form a line of sight, and then intersect at the imaginary glass. The position of each point presented on the glass is the position of the point of the three-dimensional object to be painted on the two-dimensional plane. This is the application method of perspective in Western classical painting. Perspective in a narrow sense refers specifically to the linear perspective and other scientific perspective methods that were gradually established in the 14th century to depict objects and reproduce space. In modern times, due to the study of human visual perception, the scope and content of perspective have been expanded. Perspective in a broad sense can refer to various methods of spatial expression.

[0075] The overall meaning of perspective is the technology of showing 3D stereoscopic effect on 2D plane material. In this embodiment, the position of each key point, line ratio, curvature, light and other elements are used as basic 3D framework data. The conversion module calculates these data, and the result is the basic framework data of the 3D image. According to the setting, these data are calculated to display the 3D effect, that is, the 3D image framework is calculated. After obtaining the 3D image framework, the content between each key point, line, and curvature during object conversion needs to calculate the perspective data and post-effect map filling according to the corresponding content of the 2D image.

[0076] In a possible implementation, the number of the objects to be converted is multiple;

[0077] After obtaining the first multi-angle cropped image of each object to be converted, the method further includes:

[0078] The first multi-angle cropped images of the objects to be converted are spliced ​​or the first multi-angle cropped images of the objects to be converted are spliced ​​with a preset background image to perform 3D conversion on the spliced ​​objects to be converted to obtain 3D images of the objects to be converted.

[0079] The present invention can also merge two or more 3D images to convert into a new 3D image, extract elements from a certain image and merge them with another image to convert into a 3D image to enhance the integrity of the 3D image. For equipment that is too large in size, the present invention can divide it into several parts and splice images from various angles before modeling to convert into a complete three-dimensional image; the present invention makes it possible to convert large equipment into three-dimensional images.

[0080] Generally, people are used to placing objects in the center of the viewfinder or screen when taking photos, which creates the basic conditions for AI to recognize objects. The present invention judges the shape and outline of the central object, as well as the details of the morphological state within the shape and outline, compares the central objects of several 2D images, recognizes whether it is a human or an object, cuts the human or object out of the 2D image according to the outline data, that is, the corresponding shooting data, converts the outlines of the same person or object of several 2D images, that is, the corresponding shooting data, into 3D, and combines them into a 3D person or object, which has a 360-degree dimension.

[0081] In a possible implementation, when the image source is pre-stored images of each object to be identified at multiple angles stored in a memory;

[0082] The shape outline of the object to be identified in the pre-stored images at each angle of each object to be identified and located at the center of the current shooting picture is identified and cropped to obtain a second multi-angle cropped image of each object to be identified; based on the second multi-angle cropped image of each object to be converted, each object to be converted is 3D converted to obtain a 3D image of each object to be converted.

[0083] In a possible implementation, after acquiring a multi-angle image of at least one object to be converted through a shooting terminal, the method further includes:

[0084] Eliminate distortion of multi-angle images of each object to be transformed.

[0085] In order to better display the 3D image effect, the software and hardware of the shooting equipment should have the ability to eliminate distortion. The existing shooting equipment cannot eliminate distortion and can convert 3D images, but the displayed 3D image effect will also have a distortion effect. The distortion can be modified to achieve a state acceptable to the normal human eye.

[0086] Both the shooting function and the conversion function can have the functions of automatically removing background people or debris, blurring the background, enhancing the subject, etc.

[0087] When converting a multi-angle image into a 3D image, generally no data other than the shooting data is added. The present invention can add other data such as text, symbols, images, etc. after the modeling is completed.

[0088] In a possible implementation, after obtaining the 3D image of each object to be converted, the method further includes:

[0089] The 3D image of each object to be converted is output to display the 3D image of each object to be converted.

[0090] The converted images at various angles are saved in the designated folder of the image conversion module, and the cover of the 3D image is displayed in the gallery of the image conversion module through the display device, and a 3D icon or text is displayed on the cover to distinguish the plane images at various angles.

[0091] When the user selects a 3D image in the gallery, the first frame of the 3D image displayed on the screen is the frame of the first photo or the first shot of the video.

[0092] When watching a 3D image, after the 3D image appears on the screen, you can use your fingers or a mouse to slide the image to rotate the 3D image, thereby displaying a 360-degree or multi-angle panoramic picture of the 3D image, so that the 3D image of the subject is truly presented on the screen, showing a visual experience with a sense of distance and space. The display and viewing are continuous and natural while sliding the screen image, rather than a separate 3D display of a certain image. It is as if the subject is actually in front of you and the viewer is watching around the subject.

[0093] Depending on the settings of the 3D conversion at the time, one can show that the central object and the background rotate together, and the other can show that the central object rotates but the background does not rotate.

[0094] When a 3D image is displayed in a gallery or a program storing images, 3D text or logo is added to distinguish it from a flat image in the gallery or a program storing images.

[0095] When a 3D image is displayed, the image can be viewed from the side of the screen with a 3D effect. When the screen is turned sideways, the image can also be viewed with a 3D effect.

[0096] When displaying 3D images, viewers can zoom in and out of the 3D images, and they can be cropped according to user needs.

[0097] The 3D image should have the function of marking the size of objects, and through the lens and algorithm, the image conversion module can accurately measure the distance, size, etc., and mark them in the 3D image as needed, or measure the object size in advance and manually mark it in the 3D image.

[0098] You can use dedicated image viewing software or general image viewing software to view 3D effects.

[0099] The 3D images in the present invention include static 3D images and dynamic 3D images, wherein the static 3D images are obtained based on multi-angle static images; and the dynamic 3D images are obtained based on dynamic multi-angle dynamic images.

[0100] The characters, objects or environments displayed in static 3D images are without motion, without changes in the distance between the photographer and the subject, and without changes in ambient light.

[0101] The characters, objects or environments displayed in dynamic 3D images are in motion, with changes in distance and changes in ambient light.

[0102] When there is no full-angle shooting display, the maximum rotation angle of the 3D image is the continuous shooting angle.

[0103] The shooting terminal in the present invention can be a terminal such as a computer, a camera, a camcorder, and a mobile phone; the image conversion module can be a control module embedded in the shooting terminal, or a terminal device such as a computer with image processing capabilities, and the present invention does not limit this.

[0104] Computers include desktop computers, laptops, video processing equipment, and software. Cameras include cameras of various specifications with sensors of various sizes, and cameras for mobile phones, tablets, wearable devices, and computers. Video cameras include cameras of various specifications with sensors of various sizes, and cameras for mobile phones, tablets, and wearable devices.

[0105] The image source involved in the present invention is 2D photos, pictures, images, videos taken by mobile phones, tablet computers, wearable devices, cameras, camcorders or photos, pictures, images, videos obtained from other channels.

[0106] like Figure 4 As shown, the image conversion module suitable for implementing the above-mentioned embodiment includes a central processing module (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) or the program loaded from the storage part into the random access memory (RAM). In the RAM, various programs and data required for the operation of the computer system are also stored. The CPU, ROM and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0107] The following components are connected to the I / O interface: an input part including a keyboard, a mouse, etc.; an output part including a liquid crystal display (LCD), etc. and a speaker, etc.; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a LAN card, a modem, etc. The communication part performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., are installed on the drive as needed so that the computer program read therefrom is installed into the storage part as needed.

[0108] In particular, according to this embodiment, the process described in the flowchart above can be implemented as a computer software program. For example, this embodiment includes a computer program product, which includes a computer program tangibly contained on a computer-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium.

[0109] The flow chart and schematic diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product of the present embodiment. In this regard, each square box in the flow chart or schematic diagram can represent a part of a module, program segment or code, and a part of the above-mentioned module, program segment or code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the schematic diagram and / or flow chart, and the combination of the square boxes in the schematic diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0110] In the description of the present disclosure, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present disclosure and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present disclosure. Unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, and it can be the internal connection of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present disclosure can be understood according to the specific circumstances.

[0111] It should also be noted that, in the description of the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the statement "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment including the elements.

[0112] Obviously, the above embodiments of the present disclosure are merely examples for clearly illustrating the present disclosure, and are not limitations on the implementation methods of the present disclosure. For ordinary technicians in this field, other different forms of changes or modifications can be made on the basis of the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present disclosure are still within the protection scope of the present disclosure.

Claims

1. A method for converting a two-dimensional image into a three-dimensional image, characterized in that: include: Obtain the image source required for 3D conversion through the shooting terminal; When the image source is real-time shooting, the object in the current shooting picture is positioned through the shooting terminal, and the multi-angle image of at least one object to be converted and the positioning result corresponding to each angle image of each object to be converted are obtained through the shooting terminal; According to the positioning results corresponding to the images at each angle in each object to be converted, the images at each angle in each object to be converted are respectively cropped to obtain first multi-angle cropped images of each object to be converted; Based on the first multi-angle cropped images of each object to be converted, 3D conversion is performed on each object to be converted to obtain a 3D image of each object to be converted.

2. The method according to claim 1, characterized in that The positioning of the object in the current shooting picture by the shooting terminal includes Identify the object at the center of the current shooting picture or identify the shape outline of the object in the user-set area in the current shooting picture in response to the user's setting operation, and use the shape outline of the object as the positioning result of the object in the current shooting picture.

3. The method according to claim 2, characterized in that After positioning the object in the current shooting picture by the shooting terminal, the method further includes: Identify the object at the center of the current shooting picture or respond to the user's setting operation to identify the surface material, surface color and surface light of the object in the user-set area in the current shooting picture, and judge and output the type of object in the current shooting picture based on the object's shape outline, surface material, surface color and surface light.

4. The method according to claim 2, characterized in that: The step of cutting the images at each angle of each object to be converted according to the positioning results corresponding to the images at each angle of each object to be converted comprises: The images at each angle of each object to be converted are respectively cut according to the shape contours corresponding to the images at each angle of each object to be converted; After obtaining multi-angle images of at least one object to be converted and positioning results corresponding to each angle image of each object to be converted through the shooting terminal, the method further includes: Evaluate the quality of each image at each angle of each object to be converted and output the evaluation results.

5. The method according to claim 1, characterized in that When the image source is real-time shooting, the method further includes acquiring and outputting a plurality of shooting heights and a maximum relative rotation angle required for two adjacent images taken by the shooting terminal relative to the object to be converted at each shooting height through the shooting terminal; Responding to the nth shooting operation of the user, acquiring the nth image, and identifying whether the relative rotation angle between the nth image shot by the shooting terminal relative to the object to be converted at the current shooting height and the current shooting picture shot by the shooting terminal relative to the object to be converted is less than the maximum relative rotation angle, and outputting the recognition result; Wherein n is an integer greater than 1; The maximum relative rotation angle is greater than 0° and less than or equal to 180°.

6. The method according to claim 5, characterized in that The method of identifying the relative rotation angle between the nth image captured by the shooting terminal relative to the object to be converted at the current shooting height and the current shooting image captured by the shooting terminal relative to the object to be converted comprises: The shooting terminal is fixed at a corresponding shooting height, and the object to be converted rotates around its longitudinal central axis. The shooting terminal obtains the rotation angle to obtain the nth image taken by the shooting terminal relative to the object to be converted at the current shooting height and the relative rotation angle when the shooting terminal takes the current shooting picture relative to the object to be converted; Alternatively, the object to be converted is fixed, the shooting terminal is rotated with the object to be converted as the central axis, and the shooting terminal identifies the nth image captured by the shooting terminal around the object to be converted at the current shooting height and the relative rotation angle of the shooting terminal relative to the object to be converted when the current shooting picture is captured.

7. The method according to claim 1, characterized in that The number of the objects to be converted is multiple; After obtaining the first multi-angle cropped image of each object to be converted, the method further includes: The first multi-angle cropped images of the objects to be converted are spliced ​​or the first multi-angle cropped images of the objects to be converted are spliced ​​with a preset background image to perform 3D conversion on the spliced ​​objects to be converted to obtain 3D images of the objects to be converted.

8. The method according to claim 1, characterized in that When the image source is pre-stored multi-angle images of each object to be identified stored in the memory; Recognize the shape outline of the object to be identified in the pre-stored images at each angle of each object to be identified and crop it to obtain a second multi-angle cropped image of each object to be identified; Based on the second multi-angle cropped images of each object to be converted, 3D conversion is performed on each object to be converted to obtain a 3D image of each object to be converted.

9. The method according to claim 1, characterized in that: After acquiring multi-angle images of at least one object to be converted through the shooting terminal, the method further includes: Eliminate distortion of multi-angle images of each object to be transformed.

10. The method according to claim 1, characterized in that After obtaining the 3D images of the objects to be converted, the method further includes: The 3D image of each object to be converted is output to display the 3D image of each object to be converted.

Images