Pratul Srinivasan

Pratul Srinivasan is a Researcher at Google DeepMind specializing in Neural Radiance Fields (NeRF), 3D scene reconstruction, and view synthesis at the intersection of computer vision, graphics, and machine learning. He earned his PhD from the EECS Department at UC Berkeley in 2020 under Ren Ng and Ravi Ramamoorthi, with prior research at Duke University on medical computer vision under Sina Farsiu.

Research Interests: Pratul focuses on 3D reconstruction using neural fields, light field synthesis, illumination modeling, and diffusion-based 3D generation. His work addresses challenges in photorealistic rendering, real-time view synthesis, and inverse rendering with applications in astronomy and medical imaging.

Publication Trends: Recent articles emphasize real-time NeRF (e.g., Bolt3D), refractive material modeling, shadow-based illumination recovery, and cross-scale generative synthesis. Collaborations span institutions including MIT, NVIDIA, and ETH Zurich.

Scientific Awards:

• 2025 SIGGRAPH Significant New Researcher Award
• 2021 ACM Doctoral Dissertation Award Honorable Mention
• 2020 David J. Sakrison Memorial Prize
• Best Paper Awards at ECCV 2020, ICCV 2021, and CVPR 2022

Advising & Collaborations: Advised by Ren Ng and Ravi Ramamoorthi during his PhD, Pratul collaborates with researchers like Jonathan T. Barron, Ben Mildenhall, and Katherine L. Bouman. His work integrates differentiable simulations, Fourier feature networks, and multiplane image extrapolation.

Technical Contributions: Key innovations include anti-aliased NeRF, memory-efficient rendering, refractive relighting, and multi-view relighting techniques, with impacts on virtual reality, astronomical imaging, and 3D content creation.