Bio
I lead a product-oriented research team in AI for Experiences (AI4EX) at NVIDIA. Our research is centered around AI, computer graphics, and visual perception. With collaborators, I have contributed to advances in deep learning for real-time graphics—including work on DLSS 4, RTX HDR, RTX Dynamic Vibrance, and DLSS 1—as well as perceptual metrics for image quality (FovVideoVDP), foveated rendering, and redirected walking in VR.
Previously, I was a Principal Research Scientist at NVIDIA (2021–2023), a Research Scientist at Facebook Reality Labs (2019–2021), and a Research Scientist and later Senior Research Scientist in Real-Time Rendering at NVIDIA (2013–2019). I received my Ph.D. from UC Davis in 2013 and my B.Tech. from IIT Delhi in 2007.
I also track SIGGRAPH trends.
Selected Impact
Co-developed products, coauthored research, and shared recognition
-
Product Impact
-
Research Innovations
-
Recognition
Publications & Presentations
2026
 |
Real-Time Neural Filtering for Perceptual Enhancement of Digital ExperiencesInvited Paper SID Symposium Digest of Technical Papers Paper Slides |
 |
SmoothDiffusion-VE: Real-time Generative Video Editing Using Adaptive Feature Cache IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Paper PDF |
2025
 |
Detection of artifacts in clean and corrupted video pairs is influenced by artifact type
and presentation modality Journal of Vision 25(9), VSS abstract 2256 JOV |
2023
 |
The Shortest Route Is Not Always the Fastest: Probability-Modeled Stereoscopic Eye Movement Completion Time in VR ACM Transactions on Graphics 42(6) (SIGGRAPH Asia) Project Preprint |
 |
Online Overexposed Pixels Hallucination in Videos with Adaptive Reference Frame Selection arXiv:2308.15462 arXiv |
 |
Power, Performance, and Image Quality Tradeoffs in Foveated Rendering IEEE Conference on Virtual Reality and 3D User Interfaces (VR) IEEE Preprint |
2022
 |
Modeling and Optimizing Human-in-the-Loop Visual Perception Using Immersive Displays: A ReviewInvited Paper SID Symposium Digest of Technical Papers SID Preprint |
 |
Image Features Influence Reaction Time: A Learned Probabilistic Perceptual Model for Saccade LatencyBest Paper Award SIGGRAPH Project Preprint Video |
 |
Learning to compress videos without computing motion Signal Processing: Image Communication ScienceDirect |
 |
FOVQA: Blind Foveated Video Quality Assessment IEEE Transactions on Image Processing 31, 4571–4584 IEEE Preprint |
2021
 |
FovVideoVDP: A Visible Difference Predictor for Wide Field-of-view Video SIGGRAPH Project Preprint Presentation Code |
|
FovDots: Foveated rendering dataset Dataset |
|
 |
Subjective and Objective Quality Assessment of 2D and 3D Foveated Video Compression in Virtual Reality IEEE Transactions on Image Processing Dataset IEEE |
 |
A foveated video quality assessment model using space-variant natural scene statistics IEEE International Conference on Image Processing IEEE |
 |
Evaluating Foveated Video Quality Using Entropic Differencing Picture Coding Symposium IEEE |
|
MOVI-Codec: Deep Video Compression without Motion Picture Coding Symposium IEEE |
2020
 |
Semi-Supervised StyleGAN for Disentanglement Learning ICML arXiv Slides Data |
 |
Neural Temporal Adaptive Sampling and Denoising Eurographics Project Preprint Video Recording |
 |
State of the Art in Perceptual VR Displays Book Chapter in Real VR – Immersive Digital Reality Springer |
 |
Redirected Walking in VR Book Chapter in Real VR – Immersive Digital Reality Springer |
2019
 |
Towards Deep Real-time Rendering for Mixed Reality Invited talk at Dagstuhl Seminar 19272 (Real VR - Importing the Real World into Immersive VR and Optimizing the Perceptual Experience of Head-Mounted Displays) DOI |
 |
Truly Next-Gen: Adding Deep Learning to Games & Graphics Talk at Game Developers Conference (GDC) 2019 Recording (GDC Vault) |
2018
 |
Towards Virtual Reality Infinite Walking: Dynamic Saccadic Redirection ACM Transactions on Graphics (SIGGRAPH) Project Preprint Fast Forward Video Two-minute Papers GTC Demo |
 |
Detecting Aliasing Artifacts in Image Sequences Using Deep Neural Networks High-Performance Graphics ACM Slides |
 |
Spoke-Darts for High-Dimensional Blue Noise Sampling ACM Transactions on Graphics Preprint Video Code |
 |
Applications of Vision Science to Virtual and Augmented Reality ACM SIGGRAPH 2018 Course Course Page |
2017
 |
Latency Requirements for Foveated Rendering in Virtual RealityBest Paper Award ACM Transactions on Applied Perception (Symposium on Applied Perception 2017) Project Preprint |
 |
Spatiotemporal Variance-Guided Filtering: Real-Time Reconstruction for Path-Traced Global IlluminationWolfgang Straßer Best Paper Award High-Performance Graphics 2017 Project Preprint Video |
 |
Varifocal Virtuality: A Novel Optical Layout for Near-Eye Display ACM SIGGRAPH 2017 Emerging Technologies ACM PDF |
 |
Applications of Visual Perception to Virtual Reality Rendering ACM SIGGRAPH 2017 Course Course Page |
 |
Perceptual Insights into Foveated Virtual Reality NVIDIA GTC 2017 Talk: Slides Recording SIGGRAPH 2017 Recap: Slides |
2016
 |
Towards Foveated Rendering for Gaze-Tracked Virtual Reality ACM Transactions on Graphics (SIGGRAPH Asia) Project Preprint Supplementary Material Slides Fast Forward |
 |
Perceptually-Based Foveated Virtual RealityLaval Virtual Award ACM SIGGRAPH 2016 Emerging Technologies Project Abstract Video |
 |
Filtering Distributions of Normals for Shading AntialiasingWolfgang Straßer Best Paper Award High-Performance Graphics, June 2016 Project Preprint Video |
2015
 |
Piko: A Framework for Authoring Programmable Graphics Pipelines ACM Transactions on Graphics (SIGGRAPH) ACM Preprint arXiv:1404.6293 Slides |
2014
 |
Exercises in High-Dimensional Sampling: Maximal Poisson-disk Sampling and k-d Darts Topological and Statistical Methods for Complex Data -- Tackling Large-Scale, High-Dimensional, and Multivariate Data Sets, June 2014 Springer |
 |
k-d Darts: Sampling by k-Dimensional Flat Searches ACM Transactions on Graphics ACM arXiv |
2013
 |
Programmable Graphics PipelinesZuhair A. Munir Award Honorable Mention Ph.D. Dissertation, University of California at Davis |
2012
 |
High-Quality Parallel Depth-of-Field Using Line Samples High-Performance Graphics, June 2012 ACM Slides |
 |
A Simple Algorithm for Maximal Poisson-Disk Sampling in High Dimensions Computer Graphics Forum ACM Preprint |
2011
 |
Efficient Adaptive Tiling for Programmable Rendering ACM Symposium on Interactive 3D Graphics and Games (I3D), Poster Abstract ACM |
 |
Efficient and Good Delaunay Meshes From Random Points Computer-Aided Design 43(11), 1506–1515 DOI |
 |
Efficient Maximal Poisson-Disk Sampling ACM Transactions on Graphics (SIGGRAPH) ACM Preprint |
 |
Isotropic Conforming Refinement of Quadrilateral and Hexahedral Meshes using Two-Refinement Templates International Journal for Numerical Methods in Engineering, 2011 DOI |
2010
 |
Fragment-Parallel Composite and Filter Computer Graphics Forum (EGSR 2010) DOI |
 |
Task Management for Irregular-Parallel Workloads on the GPU2019 HPG Test-of-Time Award High Performance Graphics 2010 ACM |
2009
 |
Real-Time Reyes: Analysis of a Programmable Rendering Pipeline November 2009: Talk at Crytek Conference, Frankfurt, Germany Slides |
 |
Parallel View-Dependent Tessellation of Catmull-Clark Subdivision Surfaces High Performance Graphics 2009 ACM Preprint |
2008
 |
Real-Time Reyes: Programmable Pipelines and Research Challenges Course-talk at SIGGRAPH Asia 2008, Singapore Slides |
 |
Real-Time Reyes-Style Adaptive Surface Subdivision ACM Transactions on Graphics (SIGGRAPH Asia): ACM Preprint Talk at UC Berkeley Graphics Lunch: Slides Talk at Microsoft Research: Slides |
 |
Texture Filter Memory - A Power-efficient and Scalable Texture Memory Architecture for Mobile Graphics Processors IEEE/ACM International Conference on Computer-Aided Design (ICCAD), November 2008 IEEE Xplore |
 |
Efficient computation of sum-products on GPUs through software-managed cache ACM International Conference on Supercomputing, June 2008 ACM Preprint |