Advances in real-time rendering in games

Tatarchuk, Natalya; Dupuy, Jonathan; Deliot, Thomas; Wright, Daniel B.; Narkowicz, Krzysztof; Kelly, Patrick J.; Netzel, Aleksander; Costa, Tiago

doi:10.1145/3532720.3546895

Cited by 3 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A note to SDF practitioners: tracing SDF is not necessarily faster than HWRT (as we have also demonstrated). We quote the following text from the design document of Lumen [TDD*22] as one of the important reason to use DF: “Hardware Ray Tracing is great and it is the future, but we need options to scale down. In the PC market there are still plenty of video cards that don't support hardware ray tracing, and console Hardware Ray Tracing is not that fast”.…”

Section: Resultsmentioning

confidence: 99%

“…This is commonly referred to as distance fields (DFs). However, sphere tracing the distance fields are still too slow for long and incoherent rays (computed with different numbers of iterations per thread) [TDD*22], especially for grazing angle rays shown in Figure 3(d). When rays are close to grazing angles, the sphere tracing method has to take a large number of small steps to advance the ray and therefore result in a large number of iterations.…”

Section: Background and Problem Analysismentioning

confidence: 99%

“…Distance fields (DFs) is another form of scene representation widely used in computer graphics. It can be used for ray tracing [TDD*22; LM22], in which case it is an approximation to the scene geometries and typically generated using jump flooding algorithm (JFA) [RT06] with occupancy maps as seeds. Sphere tracing is used to perform fast tracing against DFs to find nearest intersection.…”

Section: Related Workmentioning

confidence: 99%

“…Besides, distance fields are costly to generate. In real‐time applications, distance fields are typically generated either from OMs using jump flooding algorithm (JFA) [RT06; LM22] on the fly, or by closest point queries [TDD*22] in a pre‐computation manner. However, considering the cost of these methods, and the time required to filling a 3D array of floating numbers, distance fields are more expensive to support per‐frame update for dynamic objects than OMs.…”

Section: Background and Problem Analysismentioning

confidence: 99%

“…Another popular alternative is distance fields (DFs) coupled with sphere tracing [TDD*22]. DFs provide a good approximation to the scene geometry, and sphere tracing enables fast ray tracing against DFs.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Ray‐aligned Occupancy Map Array for Fast Approximate Ray Tracing

Zeng

Wang

et al. 2023

Computer Graphics Forum

View full text Add to dashboard Cite

We present a new software ray tracing solution that efficiently computes visibilities in dynamic scenes. We first introduce a novel scene representation: ray‐aligned occupancy map array (ROMA) that is generated by rasterizing the dynamic scene once per frame. Our key contribution is a fast and low‐divergence tracing method computing visibilities in constant time, without constructing and traversing the traditional intersection acceleration data structures such as BVH. To further improve accuracy and alleviate aliasing, we use a spatiotemporal scheme to stochastically distribute the candidate ray samples. We demonstrate the practicality of our method by integrating it into a modern real‐time renderer and showing better performance compared to existing techniques based on distance fields (DFs). Our method is free of the typical artifacts caused by incomplete scene information, and is about 2.5×–10× faster than generating and tracing DFs at the same resolution and equal storage.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Background and Problem Analysismentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Background and Problem Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Ray‐aligned Occupancy Map Array for Fast Approximate Ray Tracing

Zeng

Wang

et al. 2023

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

Neural Temporal Denoising for Indirect Illumination

Zeng¹,

Wang²,

Xu³

et al. 2023

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Random-Access Neural Compression of Material Textures

Vaidyanathan,

Salvi,

Wronski

et al. 2023

ACM Trans. Graph.

View full text Add to dashboard Cite

The continuous advancement of photorealism in rendering is accompanied by a growth in texture data and, consequently, increasing storage and memory demands. To address this issue, we propose a novel neural compression technique specifically designed for material textures. We unlock two more levels of detail, i.e., 16× more texels, using low bitrate compression, with image quality that is better than advanced image compression techniques, such as AVIF and JPEG XL. At the same time, our method allows on-demand, real-time decompression with random access similar to block texture compression on GPUs, enabling compression on disk and memory. The key idea behind our approach is compressing multiple material textures and their mipmap chains together, and using a small neural network, that is optimized for each material, to decompress them. Finally, we use a custom training implementation to achieve practical compression speeds, whose performance surpasses that of general frameworks, like PyTorch, by an order of magnitude.

show abstract

Advances in real-time rendering in games

Cited by 3 publications

References 0 publications

Ray‐aligned Occupancy Map Array for Fast Approximate Ray Tracing

Ray‐aligned Occupancy Map Array for Fast Approximate Ray Tracing

Neural Temporal Denoising for Indirect Illumination

Random-Access Neural Compression of Material Textures

Contact Info

Product

Resources

About