Description
Neural rendering is an emerging field at the intersection of computer graphics and machine learning. It explores how neural networks can be used to represent, compress, and reproduce complex visual phenomena that are difficult to capture with traditional techniques. Applications range from material appearance and lighting to geometry, animation, and volumetric effects.
The goal of this internship is to investigate neural rendering techniques in the context of real-time graphics (neural materials, compression, lighting, etc.). During the internship, the student will survey recent advances in the field, identify key limitations in existing methods, and implement experimental prototypes. The work is exploratory in nature, with emphasis on deriving insights that can inform future research and integration into real-time rendering systems.
References:
- T. Zeltner et al., “Real-Time Neural Appearance Models”. ACM SIGGRAPH (2024).
- A. Kuznetsov et al., “Rendering Neural Materials on Curved Surfaces”. ACM SIGGRAPH (2022).
- K. Vaidyanathan, et al., “Random-Access Neural Compression of Material Textures”, ACM SIGGRAPH (2023).
- C. Weinreich, et al., “Real-Time Neural Materials using Block-Compressed Features”. Eurographics (2024).
- A. Weidlish. “Bridging the Gap Between Offline and Real Time with Neural Materials”. Siggraph Courses (2025)
- Neural Shading (siggraph cours 2025) https://github.com/shader-slang/neural-shading-s25