Real-time rendering optimization method based on ray tracing

Abstract

The invention discloses a real-time rendering optimization method based on ray tracing. According to the method, technologies such as optimization of a spatial data structure, a light beam tracking technology, adaptive sampling, a denoising technology, illumination mapping pre-calculation, shader optimization, light-scene intersection acceleration, light source cutting, importance sampling, hardware acceleration, software-hardware cooperation, machine learning prediction, a mixed rendering technology, asynchronous calculation and dynamic resolution adjustment are combined; therefore, the calculation efficiency of ray tracing is improved, the requirement of the ray tracing for hardware performance is reduced, and the real-time or near-real-time rendering speed is realized on the premise that the image quality is not sacrificed.

Description

Technical Field

[0001] The present invention relates to the field of rendering technology, and in particular to a real-time rendering optimization method based on ray tracing. Background Art

[0002] With the improvement of computing power and the development of graphics processing technology, ray tracing has become an important means of rendering high-quality images. However, due to the complexity of ray tracing algorithms, its application in the field of real-time rendering is limited. Existing real-time rendering technologies usually rely on rasterization methods and cannot fully utilize the highly realistic visual effects brought by ray tracing.

[0003] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a real-time rendering optimization method based on ray tracing, improve the computational efficiency of ray tracing, and reduce its requirements for hardware performance, so as to achieve real-time or near real-time rendering speed without sacrificing image quality.

[0005] The technical solution of the present invention is as follows: a real-time rendering optimization method based on ray tracing is provided, comprising the following steps.

[0006] S1: Spatial data structure optimization; build efficient spatial data structures, such as bounding volume hierarchy (BVH) or kd-tree, to quickly determine the intersection of rays and objects and reduce unnecessary calculations.

[0007] S2: Ray bundle tracing; group rays into bundles and use the SIMD parallelism of the GPU for batch processing to improve computational efficiency.

[0008] S3: Adaptive sampling; dynamically adjusts the sampling rate based on the importance of the image area, applying higher sampling rates to important areas such as edges and high-contrast areas.

[0009] S4: Denoising; Apply denoising algorithms, such as Temporal Anti-Aliasing (TAA), to reduce noise in real-time rendering.

[0010] S5: Lightmap precomputation; precompute lightmaps when the scene is loaded and store lighting information for use in real-time rendering.

[0011] S6: Shader optimization: Optimize shader algorithms, reduce branching and division operations, and improve shader operation efficiency.

[0012] S7: Accelerate ray-scene intersection; use techniques such as conservative rasterization to quickly determine potential intersecting geometry and reduce the amount of ray tracing calculations.

[0013] S8: Light source clipping; analyze the visibility of light sources in the scene, ignore obscured light sources, and reduce unnecessary light projection.

[0014] S9: Importance sampling; perform importance analysis on light sources and give priority to light sources that contribute most to rendering.

[0015] S10: Hardware acceleration; use dedicated ray tracing hardware, such as the RT core of the NVIDIA RTX GPU, to accelerate ray tracing calculations. The RT core is a hardware unit designed specifically for accelerating ray tracing calculations. It accelerates the calculation of tree-based ray tracing structures, such as the bounding volume hierarchy (BVH), by quickly determining where the light intersects with the scene's 3D geometry. Compared with traditional methods that rely on general-purpose processing units for software simulation, the RT core can significantly improve the performance of ray tracing, thereby achieving real-time or near real-time rendering speeds. In this step, the dedicated ray tracing hardware is any one of the NVIDIA RTX series graphics cards, AMD RDNA series graphics cards, Intel A770 graphics cards, and Imagination Technologies' IMG CXT GPU.

[0016] S11: Software-hardware collaboration; rationally distribute tasks between the CPU and GPU, use the CPU for complex calculations and data management, and the GPU focuses on ray tracing calculations.

[0017] S12: Machine learning prediction: Use machine learning algorithms to predict the interaction between light and scenes and optimize rendering parameters.

[0018] S13: Hybrid rendering technology; combines ray tracing and rasterization rendering technologies to handle different visual effects respectively.

[0019] S14: Asynchronous calculation; performs ray tracing calculations in a background thread to reduce the impact on the main thread.

[0020] S15: Dynamic resolution and quality adjustment; dynamically adjust rendering resolution and quality based on scene dynamics and camera movement.

[0021] Furthermore, in step S1, the method for optimizing the data structure includes at least one of the following:

[0022] Pre-traversal optimization: Before traversing the tree, some information can be pre-calculated, such as pre-calculating some possible intersections, to reduce the amount of calculation during the actual traversal.

[0023] Multi-level BVH: Using a multi-level BVH, where each level has a bounding volume of a different size, can better handle objects of different sizes and shapes.

[0024] Adaptive construction: According to the characteristics of the scene and the distribution of objects, adaptively choose to build BVH or kd tree, or mix the two in different areas.

[0025] Parallel construction: Use multi-core processors to build spatial data structures in parallel to reduce preprocessing time.

[0026] Dynamic Updates: For dynamic scenes, use spatial data structures that can be updated quickly or rebuilt periodically to adapt to scene changes.

[0027] Further, in step S2, the batch processing scheme includes at least one of the following:

[0028] SIMD parallelism: Graphics processing units have a large number of SIMD units, each of which can process multiple data under a single instruction; in ray bundle tracing, all rays in the bundle can be processed synchronously in the same SIMD unit.

[0029] Intersection calculation: Intersection calculations are performed simultaneously on all rays in a ray bundle, allowing many intermediate results to be shared, such as the traversal of the bounding volume hierarchy tree.

[0030] Furthermore, in step S3, the implementation method of adaptive sampling includes at least one of the following:

[0031] Layered Sampling: During the rendering process, a multi-layer sampling strategy is adopted; first, a lower base sampling rate is used for preliminary rendering, and then based on the results of the preliminary rendering, more advanced sampling is performed on important areas.

[0032] Iterative convergence: Adaptive sampling can be used as part of an iterative process, where the importance of the region is re-evaluated after each iteration based on the current rendering results, and the sampling rate is adjusted accordingly.

[0033] Further, in step S4, a Temporal Anti-Aliasing (TAA) algorithm is used to implement denoising, which specifically includes at least one of the following:

[0034] Camera jitter: Temporal Anti-Aliasing (TAA) can capture more sub-pixel information by introducing small random offsets to the camera position or projection matrix in each frame; this jitter is particularly effective in static scenes and can significantly improve the sampling rate.

[0035] Reprojection: For dynamic scenes, Temporal Anti-Aliasing (TAA) uses reprojection technology, which reversely maps the pixels of the current frame to the positions of the previous frame according to the motion of objects and cameras to obtain historical sampling information.

[0036] Motion vector: In order to accurately match the current pixel with the pixel of the previous frame, Temporal Anti-Aliasing (TAA) uses motion vectors to compensate for the motion of objects and cameras to ensure correct superposition of temporal information.

[0037] Further, in step S5, the light map pre-calculation includes at least one of the following:

[0038] Multi-threaded computing: Use multi-core CPUs to calculate light maps in parallel, shortening pre-calculation time.

[0039] On-demand loading: Dynamically load and unload lightmaps based on the current view angle and object visibility to reduce memory usage.

[0040] Further, in step S6, the algorithm for optimizing the shader includes at least one of the following:

[0041] Reduce branches: (1) Avoid using if-else statements, as they can lead to inefficient GPU execution, especially in dynamic branching situations. (2) Avoid branches by pre-calculating or redesigning algorithms, or use compile-time constants as branch conditions. (3) Use mathematical functions and interpolation functions to replace conditional branches; for example, use step functions and mix functions to replace if-else statements, which can convert branch logic into mathematical operations and reduce branch operations.

[0042] Reduce division operations: Use pre-calculation and lookup tables to replace real-time calculations. For example, for complex calculations such as trigonometric functions, they can be pre-calculated and stored in a lookup table, and the results can be directly obtained through indexes to avoid the overhead of real-time calculations.

[0043] Further, in step S12, the specific steps are: S121: application of deep learning model; using deep learning models such as convolutional neural networks or recurrent neural networks to analyze objects and material properties in the scene, and predict lighting effects and reflection / refraction behaviors; S122: preparation of training data; in order to train these models, a large amount of labeled data is required, including light interaction effects in different scenes and corresponding rendering parameter settings; these data can be generated by existing ray tracing algorithms.

[0044] Further, in step S13, the hybrid rendering technology includes at least one of the following:

[0045] Direct lighting processing: Use rasterization rendering to process direct lighting, and use shaders to calculate the light intensity and color on the surface of objects; this method is fast and suitable for real-time rendering needs.

[0046] Indirect Lighting and Global Illumination: Ray tracing techniques are used to calculate indirect lighting, such as reflections and refractions, as well as global illumination effects; this improves the lighting realism of your scenes while maintaining computational efficiency.

[0047] By adopting the above scheme, the present invention provides a real-time rendering optimization method based on ray tracing, which has the following technical effects:

[0048] (1) Improve image quality: Through the optimization of ray tracing algorithms and the application of hardware acceleration, more realistic light and shadow effects can be achieved, presenting higher quality images in real-time rendering.

[0049] (2) Improve rendering speed: By optimizing the rendering pipeline and computing resources, as well as the adaptive sampling strategy, the rendering speed is improved. High-quality rendering results can be generated quickly without sacrificing image quality.

[0050] (3) Enhance the practicality of real-time rendering applications: Due to the improvement of rendering effect and rendering speed, this solution can be widely used in fields such as virtual reality, games, and film and television production, improving the practicality and user experience of real-time rendering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0052] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0053] See also Figure 1 The present invention provides a real-time rendering optimization method based on ray tracing, comprising the following steps.

[0054] S1: Spatial data structure optimization; build efficient spatial data structures, such as bounding volume hierarchy (BVH) or kd-tree, to quickly determine the intersection of rays and objects and reduce unnecessary calculations.

[0055] S2: Ray bundle tracing; group rays into bundles and use the SIMD parallelism of the GPU for batch processing to improve computational efficiency.

[0056] S3: Adaptive sampling; dynamically adjusts the sampling rate based on the importance of the image area, applying higher sampling rates to important areas such as edges and high-contrast areas.

[0057] S4: Denoising; Apply denoising algorithms, such as Temporal Anti-Aliasing (TAA), to reduce noise in real-time rendering.

[0058] S5: Lightmap precomputation; precompute lightmaps when the scene is loaded and store lighting information for use in real-time rendering.

[0059] S6: Shader optimization: Optimize shader algorithms, reduce branching and division operations, and improve shader operation efficiency.

[0060] S7: Accelerate ray-scene intersection; use techniques such as conservative rasterization to quickly determine potential intersecting geometry and reduce the amount of ray tracing calculations.

[0061] S8: Light source clipping; analyze the visibility of light sources in the scene, ignore obscured light sources, and reduce unnecessary light projection.

[0062] S9: Importance sampling; perform importance analysis on light sources and give priority to light sources that contribute most to rendering.

[0063] S10: Hardware acceleration; use dedicated ray tracing hardware, such as the RT core of the NVIDIA RTX GPU, to accelerate ray tracing calculations. The RT core is a hardware unit designed specifically for accelerating ray tracing calculations. It accelerates the calculation of tree-based ray tracing structures, such as the bounding volume hierarchy (BVH), by quickly determining where the light intersects with the scene's 3D geometry. Compared with traditional methods that rely on general-purpose processing units for software simulation, the RT core can significantly improve the performance of ray tracing, thereby achieving real-time or near real-time rendering speeds. In this step, the dedicated ray tracing hardware is any one of the NVIDIA RTX series graphics cards, AMD RDNA series graphics cards, Intel A770 graphics cards, and Imagination Technologies' IMG CXT GPU.

[0064] S11: Software-hardware collaboration; reasonably distribute tasks between the CPU and GPU, use the CPU for complex calculations and data management, and the GPU focuses on ray tracing calculations.

[0065] S12: Machine learning prediction: Use machine learning algorithms to predict the interaction between light and scenes and optimize rendering parameters.

[0066] S13: Hybrid rendering technology; combines ray tracing and rasterization rendering technologies to handle different visual effects respectively.

[0067] S14: Asynchronous calculation; performs ray tracing calculations in a background thread to reduce the impact on the main thread.

[0068] S15: Dynamic resolution and quality adjustment; dynamically adjust rendering resolution and quality based on scene dynamics and camera movement.

[0069] In this embodiment, in step S1, the method for optimizing the data structure includes at least one of the following:

[0070] Pre-traversal optimization: Before traversing the tree, some information can be pre-calculated, such as pre-calculating some possible intersections, to reduce the amount of calculation during the actual traversal.

[0071] Multi-level BVH: Using a multi-level BVH, where each level has a bounding volume of a different size, can better handle objects of different sizes and shapes.

[0072] Adaptive construction: According to the characteristics of the scene and the distribution of objects, adaptively choose to build BVH or kd tree, or mix the two in different areas.

[0073] Parallel construction: Use multi-core processors to build spatial data structures in parallel to reduce preprocessing time.

[0074] Dynamic Updates: For dynamic scenes, use spatial data structures that can be updated quickly or rebuilt periodically to adapt to scene changes.

[0075] In this embodiment, in step S2, the batch processing scheme includes at least one of the following:

[0076] SIMD parallelism: Graphics processing units have a large number of SIMD units, each of which can process multiple data under a single instruction; in ray bundle tracing, all rays in the bundle can be processed synchronously in the same SIMD unit.

[0077] Intersection calculation: Intersection calculations are performed simultaneously on all rays in a ray bundle, allowing many intermediate results to be shared, such as the traversal of the bounding volume hierarchy tree.

[0078] In this embodiment, in step S3, the implementation method of adaptive sampling includes at least one of the following:

[0079] Layered Sampling: During the rendering process, a multi-layer sampling strategy is adopted; first, a lower base sampling rate is used for preliminary rendering, and then based on the results of the preliminary rendering, more advanced sampling is performed on important areas.

[0080] Iterative convergence: Adaptive sampling can be used as part of an iterative process, where the importance of the region is re-evaluated after each iteration based on the current rendering results, and the sampling rate is adjusted accordingly.

[0081] In this embodiment, in step S4, a Temporal Anti-Aliasing (TAA) algorithm is used to implement denoising, which specifically includes at least one of the following:

[0082] Camera jitter: Temporal Anti-Aliasing (TAA) can capture more sub-pixel information by introducing small random offsets to the camera position or projection matrix in each frame; this jitter is particularly effective in static scenes and can significantly improve the sampling rate.

[0083] Reprojection: For dynamic scenes, Temporal Anti-Aliasing (TAA) uses reprojection technology, which reversely maps the pixels of the current frame to the positions of the previous frame according to the motion of objects and cameras to obtain historical sampling information.

[0084] Motion vector: In order to accurately match the current pixel with the pixel of the previous frame, Temporal Anti-Aliasing (TAA) uses motion vectors to compensate for the motion of objects and cameras to ensure correct superposition of temporal information.

[0085] In this embodiment, in step S5, the light map pre-calculation includes at least one of the following:

[0086] Multi-threaded computing: Use multi-core CPUs to calculate light maps in parallel, shortening pre-calculation time.

[0087] On-demand loading: Dynamically load and unload lightmaps based on the current view angle and object visibility to reduce memory usage.

[0088] In this embodiment, in step S6, the algorithm for optimizing the shader includes at least one of the following:

[0089] Reduce branches: (1) Avoid using if-else statements, as they can lead to inefficient GPU execution, especially in dynamic branching situations. (2) Avoid branches by pre-calculating or redesigning algorithms, or use compile-time constants as branch conditions. (3) Use mathematical functions and interpolation functions to replace conditional branches; for example, use step functions and mix functions to replace if-else statements, which can convert branch logic into mathematical operations and reduce branch operations.

[0090] Reduce division operations: Use pre-calculation and lookup tables to replace real-time calculations. For example, for complex calculations such as trigonometric functions, they can be pre-calculated and stored in a lookup table, and the results can be directly obtained through indexes to avoid the overhead of real-time calculations.

[0091] In this embodiment, in step S12, the specific steps are: S121: application of deep learning model; using deep learning models such as convolutional neural networks or recurrent neural networks to analyze objects and material properties in the scene, and predict lighting effects and reflection / refraction behaviors; S122: preparation of training data; in order to train these models, a large amount of labeled data is required, including light interaction effects in different scenes and corresponding rendering parameter settings; these data can be generated by existing ray tracing algorithms.

[0092] In this embodiment, in step S13, the hybrid rendering technology includes at least one of the following:

[0093] Direct lighting processing: Use rasterization rendering to process direct lighting, and use shaders to calculate the light intensity and color on the surface of objects; this method is fast and suitable for real-time rendering needs.

[0094] Indirect Lighting and Global Illumination: Ray tracing techniques are used to calculate indirect lighting, such as reflections and refractions, as well as global illumination effects; this improves the lighting realism of your scenes while maintaining computational efficiency.

[0095] In summary, the present invention provides a real-time rendering optimization method based on ray tracing, which has the following technical effects:

[0096] (1) Improve image quality: Through the optimization of ray tracing algorithms and the application of hardware acceleration, more realistic light and shadow effects can be achieved, presenting higher quality images in real-time rendering.

[0097] (2) Improve rendering speed: By optimizing the rendering pipeline and computing resources, as well as the adaptive sampling strategy, the rendering speed is improved. High-quality rendering results can be generated quickly without sacrificing image quality.

[0098] (3) Enhance the practicality of real-time rendering applications: Due to the improvement of rendering effect and rendering speed, this solution can be widely used in fields such as virtual reality, games, and film and television production, improving the practicality and user experience of real-time rendering applications.

[0099] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A real-time rendering optimization method based on ray tracing, characterized in that: The following steps are involved: S1: Spatial data structure optimization; building an efficient spatial data structure to quickly determine the intersection of light and objects; S2: Ray bundle tracing; group rays into bundles and use the GPU's SIMD parallelism for batch processing; S3: Adaptive sampling; dynamically adjust the sampling rate according to the importance of the image area, and apply a higher sampling rate to important areas; S4: Denoising; applying a denoising algorithm to reduce noise in real-time rendering; S5: Lightmap precomputation: pre-calculate lightmaps when the scene is loaded, and store lighting information for use in real-time rendering; S6: Shader optimization: optimize the shader algorithm, reduce branch and division operations, and improve shader operation efficiency; S7: ray-scene intersection acceleration; Quickly determine potential intersecting geometries to reduce the amount of ray tracing calculations; S8: Light source clipping; analyze the visibility of light sources in the scene, ignore obscured light sources, and reduce unnecessary light projection; S9: Importance sampling; Analyze the importance of light sources and give priority to those that contribute most to rendering; S10: Hardware acceleration; using dedicated ray tracing hardware to accelerate ray tracing calculations; S11: Software-hardware collaboration; rationally distribute tasks between CPU and GPU, use CPU for complex calculations and data management, and GPU focus on ray tracing calculations; S12: Machine learning prediction; Use machine learning algorithms to predict the interaction between light and scenes and optimize rendering parameters; S13: Hybrid rendering technology; combining ray tracing and rasterization rendering technology to handle different visual effects; S14: Asynchronous computing; ray tracing calculations are performed in the background thread to reduce the impact on the main thread; S15: Dynamic resolution and quality adjustment; dynamically adjust rendering resolution and quality based on scene dynamics and camera movement.

2. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S1, the method for optimizing the data structure includes at least one of the following: Pre-traversal optimization: Before traversing the tree, some information can be pre-calculated, such as pre-calculating some possible intersections, to reduce the amount of calculation during the actual traversal; Multi-level BVH: Use a multi-level BVH, where each level has a bounding volume of a different size, to better handle objects of different sizes and shapes; Adaptive construction: Adaptively choose to build BVH or kd tree according to the characteristics of the scene and the distribution of objects, or mix the two in different areas; Parallel construction: Use multi-core processors to build spatial data structures in parallel to reduce preprocessing time; Dynamic Updates: For dynamic scenes, use spatial data structures that can be updated quickly or rebuilt periodically to adapt to scene changes.

3. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S2, the batch processing scheme includes at least one of the following: SIMD parallelism: Graphics processing units have a large number of SIMD units, each of which can process multiple data under one instruction; Intersection calculation: Intersection calculations are performed on all rays in a bundle simultaneously, allowing many intermediate results to be shared.

4. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S3, the implementation method of adaptive sampling includes at least one of the following: Layered sampling: During the rendering process, a multi-layer sampling strategy is adopted; first, a lower base sampling rate is used for preliminary rendering, and then based on the results of the preliminary rendering, more advanced sampling is performed on important areas; Iterative convergence: Adaptive sampling can be used as part of an iterative process, where the importance of the region is re-evaluated after each iteration based on the current rendering results, and the sampling rate is adjusted accordingly.

5. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S4, a Temporal Anti-Aliasing algorithm is used to implement denoising, which specifically includes at least one of the following: Camera shake: Temporal Anti-Aliasing can capture more sub-pixel information by introducing a small random offset to the camera position or projection matrix in each frame; Reprojection: For dynamic scenes, Temporal Anti-Aliasing uses reprojection technology, which reversely maps the pixels of the current frame to the positions of the previous frame according to the movement of objects and cameras to obtain historical sampling information; Motion vector: In order to accurately match the current pixel with the pixel of the previous frame, Temporal Anti-Aliasing uses motion vectors to compensate for the movement of objects and cameras to ensure the correct superposition of temporal information.

6. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S5, the light map pre-calculation includes at least one of the following: Multi-threaded computing: Use multi-core CPUs to calculate light maps in parallel, shortening pre-calculation time; On-demand loading: Dynamically load and unload lightmaps based on the current view angle and object visibility to reduce memory usage.

7. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S6, the algorithm for optimizing the shader includes at least one of the following: Reduce branches: Avoid using if-else statements; Avoid branches by precomputing or redesigning the algorithm, or use compile-time constants as branch conditions; Use math and interpolation functions to replace conditional branches; Reduce division operations: replace real-time calculations with pre-calculation and table lookup.

8. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S12, the specific steps are: S121: Application of deep learning models; Use deep learning models such as convolutional neural networks or recurrent neural networks to analyze the properties of objects and materials in the scene and predict lighting effects and reflection / refraction behaviors; S122: Preparation of training data: In order to train these models, a large amount of labeled data is required, including light interaction effects in different scenes and corresponding rendering parameter settings.

9. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S13, the hybrid rendering technology includes at least one of the following: Direct lighting processing: Use rasterization rendering to process direct lighting, and use shaders to calculate the light intensity and color on the surface of objects; Indirect lighting and global illumination: Use ray tracing technology to calculate indirect lighting and global illumination effects.

10. The real-time rendering optimization method based on ray tracing according to claim 1, characterized in that: In step S10, the dedicated ray tracing hardware is any one of NVIDIA RTX series graphics cards, AMD RDNA series graphics cards, Intel A770 graphics cards, and Imagination Technologies' IMG CXT GPU.

Images