Real-time volume smoke rendering method and system based on multi-channel hybrid rendering pipeline

Abstract

The application provides a kind of real-time volume smoke rendering method and system based on multi-channel hybrid rendering pipeline.The application uses stencil test to accurately control volume rendering area, greatly reduces invalid calculation, compared with prior art: adopt SDF and density field coupling physical interaction model: scene geometry is represented as SDF, geometry-fluid interaction is calculated in real time in volume rendering;Adaptive ray stepping optimization: dynamically adjust the step length based on SDF distance, greatly improve the stepping efficiency away from the surface;Depth-aware physically correct mixing: based on ray propagation theory, realize the correct mixing of volume and surface, solve the transparency sorting problem;Multi-channel decoupling architecture: complex problems are decomposed into simple subproblems, which is convenient for optimization and extension.

Description

Technical Field

[0001] This invention relates to the field of real-time rendering technology, specifically to a real-time volumetric smoke rendering method and system based on a multi-channel hybrid rendering pipeline. Background Technology

[0002] Volumetric rendering is an important technique in computer graphics used to simulate the visual effects of participating media (such as smoke, clouds, and flames). Traditional volumetric rendering methods mainly include:

[0003] Voxel-based methods: These methods divide the 3D space into a regular grid, store the optical properties of each voxel, and render using light projection. While producing high quality, they consume a lot of memory and are unsuitable for dynamic scenes. For example, patent US20180040123A1 discloses a GPU-based voxelized volume rendering method, but it requires pre-computation and a large amount of video memory. Particle system-based methods: These methods render a large number of semi-transparent particles to approximate volume effects. For example, patent CN110136200A discloses a smoke simulation method based on a GPU particle system, but it has high computational overhead and struggles to achieve continuous volume and complex lighting interactions. Ray stepping-based methods: These methods emit light from a camera and sample the density field at equal intervals along the light direction. Patent WO2020123456A1 discloses an adaptive ray stepping volume rendering method, but it does not solve the problem of accurate interaction between volume and scene geometry.

[0004] It is evident that current volumetric rendering technologies face challenges in practical applications, including difficulty in achieving both performance and quality, inaccurate depth sorting, inflexible region control, unrealistic physical interactions, and poor platform compatibility. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a real-time volumetric smoke rendering method and system based on a multi-channel hybrid rendering pipeline, in order to solve the technical problems faced by the prior art, such as difficulty in achieving both performance and quality, inaccurate depth sorting, inflexible region control, unrealistic physical interaction, and poor platform compatibility.

[0006] This invention provides a real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline, comprising:

[0007] S1. Construct the first rendering pass, render the opaque geometry in the scene forward, and mark the rendering area where volume effects need to be rendered.

[0008] S2. Introduce SDF scene loading to generate SDF data, gradient data, and solid mask for opaque geometry, which are used to constrain the generation and stepping of volumetric smoke.

[0009] S3. Construct a second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result;

[0010] S4 constructs a third rendering pass, which performs depth-aware blending of the forward rendering results and the volume rendering results.

[0011] Optionally, the opaque geometry in the forward-rendered scene, and the rendering area where the volumetric effect needs to be rendered, includes:

[0012] S11. Initialize the template buffer and set the template value of all pixels to 0;

[0013] S12. Enable template testing, configure it to always pass, and set the template value of pixels that pass the test to 1;

[0014] S13. Disable color and depth writing, render the volume region boundary geometry, and mark the volume region in the stencil buffer;

[0015] S14. Restore the normal template test configuration and render the opaque geometry of the entire scene;

[0016] S15, Output scene color texture and depth texture.

[0017] Optionally, the step of introducing SDF scene loading to generate SDF data and gradient data of opaque geometry includes:

[0018] S21. Export the peripheral SDF data from the SDF data and obtain the boundary of the maximum gradient value change, which is used to construct the solid boundary extension region;

[0019] S22. Extend two grids along the gradient direction outside the maximum boundary to form an outer SDF dataset to expand the transition buffer between smoke and obstacles;

[0020] S23. Convert the peripheral SDF dataset into a single-precision floating-point texture and submit it to the second rendering channel.

[0021] Optionally, within the rendering area, volumetric smoke is rendered using a ray-stepping method, and the volumetric rendering result is output, including:

[0022] S31. Enable template testing, process pixels with a template value of 1, and emit a ray from the camera position for each pixel;

[0023] S32. Calculate the interval of intersection points between the ray and the volume region. Output the color and transparency of the rendered volume, where t_min represents the distance parameter of the ray entering the volume bounding box, t_max represents the distance parameter of the ray leaving the volume bounding box, and t represents the current sampling position along the ray direction; when t is within [t_min, t_max] and the current cumulative transmittance is not lower than the preset threshold, step sampling continues.

[0024] Optionally, the interval of the intersection of the calculated ray and the volume region Output the color and transparency of the rendered volume, including:

[0025] S321, Initial Transmittance radiance ;

[0026] S322, Determine when If the condition is met, then step S323 is executed, where This represents the current sampling position along the ray direction;

[0027] S323. Start stepping from the initial sampling position, and calculate the three-dimensional coordinates corresponding to the current sampling position at each step;

[0028] S324. Sample SDF data and obtain the signed distance d from the sampling point to the nearest geometric surface;

[0029] S325. Calculate the current single density based on the density field calculation model.

[0030]

[0031] in, Indicates sampling point The smoke density at time t Represents the basic noise density field. For smooth transition function, For blocking functions, Represents sampling points ,time The flow perturbation density term related to distance d;

[0032] S326. Calculate the illumination contribution based on the current unit density, local illumination attenuation, and current cumulative transmittance.

[0033] S327. Calculate the cumulative transmittance T_acc, expressed as:

[0034]

[0035] Where T_i represents the local transmittance of the i-th sampling segment, and T_acc represents the total transmittance accumulated from the ray origin to the current sampling point;

[0036] S328. Calculate the cumulative radiance L_acc, expressed as:

[0037]

[0038] Where L_i represents the illumination contribution of the i-th sampling point, and L_acc represents the accumulated volumetric color or radiance result along the ray direction. To adaptively adjust the step size, it is represented as:

[0039] ;

[0041] S329, until satisfied The process terminates at a certain point and outputs the color L and transparency.

[0042] Optionally, the step of performing depth-aware blending of the forward rendering result and the volume rendering result includes:

[0043] S41. Disable template testing and depth testing;

[0044] S42. For full-screen quadrilateral rendering, perform composite calculations on each pixel to sample the scene color C_scene, sample the volume color C_volume, and the transparency α.

[0045] S43. Calculate the equivalent transmittance T_eq, expressed as:

[0046]

[0047] The final color C_final is calculated as follows:

[0048] C_final = T_eq × C_scene + (1 - T_eq) × C_volume.

[0050] The present invention also provides a real-time volumetric smoke rendering system based on a multi-channel hybrid rendering pipeline, for executing the real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in any of the preceding claims, comprising:

[0051] The first rendering module is used to construct the first rendering pass, render opaque geometry in the scene forward, and mark the rendering area where volumetric effects need to be rendered.

[0052] The SDF scene module is used to import SDF scene loading and generate SDF data and gradient data for opaque geometry.

[0053] The second rendering module is used to construct the second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result;

[0054] The third rendering module is used to construct the third rendering pass and perform depth-aware blending of the forward rendering results and the volume rendering results.

[0055] Compared with the prior art, the present invention:

[0056] Template testing is used to precisely control the volume rendering area, significantly reducing unnecessary calculations; scene geometry is represented as SDF, and geometry-fluid interaction is calculated in real time during volume rendering; the step size is dynamically adjusted based on SDF distance, which greatly improves stepping efficiency when moving away from the surface; at the same time, the correct blending of volume and surface is achieved based on the theory of light propagation, solving the transparency sorting problem. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0060] Figure 2 This is a rendering effect diagram of a conventional method in one embodiment of the present invention;

[0061] Figure 3 This is a rendering effect diagram of the method in one embodiment of the present invention;

[0062] Figure 4 This is a rendering effect diagram of the method in one embodiment of the present invention;

[0063] Figure 5 This is a rendering effect diagram of the method in one embodiment of the present invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other implementation cases obtained by those skilled in the art without creative effort are within the scope of protection of this application. Functional units with the same reference numerals in the examples of this invention have the same and similar structures and functions.

[0065] See Figure 1 A real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline, characterized by comprising:

[0066] S1. Construct the first rendering pass, render the opaque geometry in the scene forward, and mark the rendering area where volume effects need to be rendered.

[0067] S2. Introduce SDF scene loading to generate SDF data, gradient data, and solid mask for opaque geometry, which are used to constrain the generation and stepping of volumetric smoke.

[0068] S3. Construct a second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result;

[0069] S4 constructs a third rendering pass, which performs depth-aware blending of the forward rendering results and the volume rendering results.

[0070] In this embodiment, S1, construct the first rendering channel, render the opaque geometry in the scene forward, and mark the rendering area where the volume effect needs to be rendered;

[0071] This step primarily involves rendering opaque geometry in the scene and marking the areas where volumetric effects need to be rendered. This is achieved through the following specific steps:

[0072] S11. Initialize the template buffer and set the template value of all pixels to 0;

[0073] S12. Enable template testing, configure it to always pass, and set the template value of pixels that pass the test to 1;

[0074] S13. Disable color and depth writing, render volume region boundary geometry (such as a cube), and mark the volume region in the stencil buffer;

[0075] S14. Restore the normal template test configuration and render the opaque geometry of the entire scene;

[0076] S15, Output scene color texture (RGBA16F format) and depth texture (depth + stencil format).

[0077] This step uses a template buffer as a spatial mask to precisely control the volume rendering range and supports volume regions of arbitrarily complex shapes (by rendering corresponding geometric markers). It is decoupled from subsequent channels and can be optimized independently.

[0078] S2. Introduce SDF scene loading to generate SDF data, gradient data, and solid mask for opaque geometry, which are used to constrain the generation and stepping of volumetric smoke.

[0079] During implementation, OpenVDB software is used to directly generate SDF and gradient data for the models in the scene, generating 512*512*1028 data points for each individual model. However, this method only requires the SDF data of the model's outer perimeter, so in OpenVDB:

[0080] 1. During the data export process, only a portion of the external SDF data is exported via code:

[0081] 2. Obtain the maximum boundary of gradient value mutation (from outside to inside, SDF changes from positive to negative values, gradient value mutation).

[0082] 3. Extend two grids outwards along the gradient direction outside the maximum boundary of the mutation to form the outer SDF dataset (reducing the dataset size to 512*512*6).

[0083] 4. Convert the external SDF dataset into single-precision floating-point textures (compress the four models into one texture) and submit it to the GPU.

[0084] S3. Construct a second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result;

[0085] Volumetric smoke is rendered only within the rendering area marked by the template using a ray stepping method, and physical interaction between the smoke and the scene geometry is achieved.

[0086] S31. Enable template testing, process pixels with a template value of 1, and emit a ray from the camera position for each pixel;

[0087] S32. Calculate the interval of intersection points between the ray and the volume region. The output renders the color and transparency of the volume, where t_min represents the distance the ray enters the volume bounding box, t_max represents the distance the ray leaves the volume bounding box, and t represents the current sampling position along the ray direction; when t is within [t_min, t_max] and the current cumulative transmittance is not lower than the preset threshold, subsequent step sampling continues.

[0088] S321, Initial Transmittance radiance ;

[0089] S322, Determine when If the condition is met, then step S323 is executed, where This represents the current sampling position along the ray direction;

[0090] S323. Start stepping from the initial sampling position, and calculate the three-dimensional coordinates corresponding to the current sampling position at each step;

[0091] S324. Sample SDF data and obtain the signed distance d from the sampling point to the nearest geometric surface;

[0092] Where d is the signed distance from the sampling point to the nearest geometric surface. When d≤0, it means that the sampling point is located inside the solid and the smoke density is 0. When d is close to 0, the attenuation of smoke near the obstacle surface is controlled by a smooth transition function, and the flow around the obstacle is combined with the flow disturbance term to simulate the deflection flow of smoke along the obstacle boundary.

[0093] S325. Calculate the current single density based on the density field calculation model. , represented as:

[0094]

[0095] in, Indicates sampling point The smoke density at time t Represents the basic noise density field. For smooth transition function, For blocking functions, Represents sampling points ,time The flow perturbation density term related to distance d;

[0096] S326. Calculate the illumination contribution based on the current unit density, local illumination attenuation, and current cumulative transmittance.

[0097] S327. Calculate the cumulative transmittance T_acc, expressed as:

[0098]

[0099] Where T_i represents the local transmittance of the i-th sampling segment, and T_acc represents the total transmittance accumulated from the ray origin to the current sampling point. Both are used together to describe the degree of light attenuation by the smoke medium.

[0100] S328. Calculate the cumulative radiance L_acc, expressed as:

[0101] Where L_i represents the illumination contribution of the i-th sampling point, and L_acc represents the volumetric color or radiance result accumulated along the ray direction.

[0102]

[0103] Where L_i represents the illumination contribution of the i-th sampling point, and L_acc represents the accumulated volumetric color or radiance result along the ray direction. To adaptively adjust the step size, it is represented as:

[0104] ;

[0106] S329, until satisfied The process terminates at a certain point and outputs color L and transparency; the color L can be further written into a volumetric color texture as input for subsequent volumetric rendering or depth-aware blending.

[0107] SDF-guided adaptive ray stepping converts the scene geometry within a volume region into a directed distance field (SDF) representation and dynamically adjusts the step size based on the SDF value at the current position during ray stepping. When a sampling point enters the interior of a solid, the corresponding density is zeroed and further smoke injection is prevented. When a sampling point approaches the surface of an object, a boundary attenuation and flow deflection effect are formed through a smooth transition function and a flow perturbation term, thereby creating a more natural spatial relationship between smoke and obstacles.

[0108] The following effects can be achieved using physically correct interaction models:

[0109] 1. Blocking effect: When a ray enters the interior of the geometry (SDF<0), its density is forced to zero;

[0110] 2. Flow simulation: Add eddy current perturbations based on SDF distance gradient near the surface of the geometry;

[0111] 3. Boundary layer effect: Near the surface, density and flow velocity are affected by the SDF distance.

[0112] S4 constructs a third rendering pass, which performs depth-aware blending of the forward rendering results and the volume rendering results.

[0113] S41. Disable template testing and depth testing;

[0114] S42. For full-screen quadrilateral rendering, perform composite calculations on each pixel to sample the scene color C_scene, sample the volume color C_volume, and the transparency α.

[0115] S43. Calculate the equivalent transmittance T_eq, expressed as:

[0116] Where T_eq represents the equivalent transmittance of the volume layer to the background scene, C_scene represents the scene color obtained by forward rendering, C_volume represents the volume rendering result, and the final color is obtained by mixing C_scene and C_volume according to T_eq.

[0117]

[0118] The final color C_final is calculated as follows:

[0119] Preferably, the final color satisfies C_final = T_eq × C_scene + (1 - T_eq) × C_volume, thereby superimposing the volumetric color effect of smoke while ensuring the visibility of the background.

[0120]

[0121] It can also be expressed as:

[0122]

[0123] in, For volumetric light emission, This represents the total distance propagated within the current volume, and the volume boundary is then smoothly transitioned.

[0124] This invention is not a simple alpha blending, but rather a physically based light propagation model; it considers the actual propagation distance and attenuation of light within a volume; and it handles the correct occlusion relationship between the volume and opaque surfaces. By using stencil testing to precisely control the volume rendering area, it significantly reduces unnecessary calculations compared to existing technologies.

[0125] 1. Employ a physical interaction model that couples SDF with density field: Represent the scene geometry as SDF and calculate the geometry-fluid interaction in real time during volume rendering;

[0126] 2. Adaptive ray stepping optimization: Dynamically adjust the step size based on SDF distance to significantly improve stepping efficiency when moving away from the surface;

[0127] 3. Physically correct blending for depth perception: Achieving correct blending of volume and surface based on the theory of light propagation, thus solving the problem of transparency ranking;

[0128] 4. Multi-channel decoupled architecture: Decomposes complex problems into simple sub-problems, which facilitates optimization and expansion.

[0129] See the table below for the differences between conventional particle systems and conventional light-stepping and this invention:

[0130] Technical indicators Traditional particle systems Traditional light stepping This invention Frame rate (1080p) 30-45 FPS 15-25 FPS 60+ FPS Memory usage High (>500MB) Medium (~300MB) Low (<200MB) Visual effects Discrete, discontinuous Continuous, but without interaction Continuous, physical interaction Depth correctness Difference medium excellent Boundary fusion hard edge Adjustable transition Adaptive smoothing Object interaction Simple collision none Physically correct obstruction of flow around

[0131] Combination Figures 2-5 As can be seen, the traditional method is physically incorrect, showing smoke passing through objects with harsh boundaries, visual discontinuity, no flow effect, and lack of detail. In contrast, the smoke in this invention is physically correct when blocked by objects, with natural and smooth transitions at the boundaries, and a strong sense of realism with a flow effect.

[0132] The present invention also provides a real-time volumetric smoke rendering system based on a multi-channel hybrid rendering pipeline, for executing the real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in any of the preceding claims, comprising:

[0133] The first rendering module is used to construct the first rendering pass, render opaque geometry in the scene forward, and mark the rendering area where volumetric effects need to be rendered.

[0134] The SDF scene module is used to import SDF scene loading and generate SDF data and gradient data for opaque geometry.

[0135] The second rendering module is used to construct the second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result;

[0136] The third rendering module is used to construct the third rendering pass and perform depth-aware blending of the forward rendering results and the volume rendering results.

[0137] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0138] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline, characterized in that, include: S1. Construct the first rendering pass, render the opaque geometry in the scene forward, and mark the rendering area where volumetric effects need to be rendered. S2. Introduce SDF scene loading to generate SDF data, gradient data, and solid mask for opaque geometry, which are used to constrain the generation and stepping of volumetric smoke. S3. Construct a second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result; S4 constructs a third rendering channel, which performs depth-aware blending of the forward rendering results and the volume rendering results.

2. The real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in claim 1, characterized in that, The opaque geometry in the forward-rendered scene, and the rendering areas where volumetric effects need to be rendered, are marked, including: S11. Initialize the template buffer and set the template value of all pixels to 0; S12. Enable template testing, configure it to always pass, and set the template value of pixels that pass the test to 1; S13. Disable color and depth writing, render the volume region boundary geometry, and mark the volume region in the stencil buffer; S14. Restore the normal template test configuration and render the opaque geometry of the entire scene; S15, Output scene color texture and depth texture.

3. The real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in claim 2, characterized in that, The process of introducing SDF scene loading to generate SDF data and gradient data for opaque geometry includes: S21. Export the peripheral SDF data from the SDF data and obtain the boundary of the maximum gradient value change, which is used to construct the solid boundary extension region; S22. Extend two grids along the gradient direction outside the maximum boundary to form an outer SDF dataset to expand the transition buffer between smoke and obstacles; S23. Convert the peripheral SDF dataset into a single-precision floating-point texture and submit it to the second rendering channel.

4. The real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in claim 3, characterized in that, Within the rendering area, volumetric smoke is rendered using a ray-stepping method, and the volumetric rendering result is output, including: S31. Enable template testing, process pixels with a template value of 1, and emit a ray from the camera position for each pixel; S32. Calculate the interval of intersection points between the ray and the volume region. Output the color and transparency of the rendered volume, where t_min represents the distance parameter of the ray entering the volume bounding box, t_max represents the distance parameter of the ray leaving the volume bounding box, and t represents the current sampling position along the ray direction. When t is within [t_min, t_max] and the current cumulative transmittance is not lower than the preset threshold, step sampling continues.

5. The real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in claim 4, characterized in that, The interval of the intersection of the computational ray and the volume region Output the color and transparency of the rendered volume, including: S321, Initial Transmittance radiance ; S322, Determine when If the condition is met, then step S323 is executed, where This represents the current sampling position along the ray direction; S323. Start stepping from the initial sampling position, and calculate the three-dimensional coordinates corresponding to the current sampling position at each step; S324. Sample SDF data and obtain the signed distance d from the sampling point to the nearest geometric surface; S325. Calculate the current single density based on the density field calculation model. , is represented as: in, Indicates sampling point The smoke density at time t Represents the basic noise density field. For smooth transition function, For blocking functions, Represents sampling points ,time The flow perturbation density term related to distance d; S326. Calculate the illumination contribution based on the current unit density, local illumination attenuation, and current cumulative transmittance. S327. Calculate the cumulative transmittance T_acc, expressed as: Where T_i represents the local transmittance of the i-th sampling segment, and T_acc represents the total transmittance accumulated from the ray origin to the current sampling point; S328. Calculate the cumulative radiance L_acc, expressed as: Where L_i represents the illumination contribution of the i-th sampling point, and L_acc represents the accumulated volumetric color or radiance result along the ray direction. To adaptively adjust the step size, it is represented as: ； S329, until satisfied The process terminates at a certain point and outputs the color L and transparency.

6. The real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in claim 5, characterized in that, The process of performing depth-aware blending of the forward rendering result and the volume rendering result includes: S41. Disable template testing and depth testing; S42. For full-screen quadrilateral rendering, perform composite calculations on each pixel to sample the scene color C_scene, sample the volume color C_volume, and the transparency α. S43. Calculate the equivalent transmittance T_eq, expressed as: The final color C_final is calculated as follows: C_final = T_eq × C_scene + (1 - T_eq) × C_volume.

7. A real-time volumetric smoke rendering system based on a multi-channel hybrid rendering pipeline, used to execute the real-time volumetric smoke rendering method based on a multi-channel hybrid rendering pipeline as described in any one of claims 1 to 6, characterized in that, include: The first rendering module is used to construct the first rendering pass, render opaque geometry in the scene forward, and mark the rendering area where volumetric effects need to be rendered. The SDF scene module is used to import SDF scene loading and generate SDF data and gradient data for opaque geometry. The second rendering module is used to construct the second rendering channel, and render volumetric smoke within the rendering area using the ray stepping method, and output the volumetric rendering result; The third rendering module is used to construct the third rendering pass and perform depth-aware blending of the forward rendering results and the volume rendering results.

Images