HyperReel: High-Fidelity 6-DoF Video with Ray-Conditioned Sampling

Volumetric scene representations enable photorealistic view synthesis for static scenes and form the basis of several existing 6-DoF video techniques. However, the volume rendering procedures that drive these representations necessitate careful trade-offs in terms of quality, rendering speed, and memory efficiency. In particular, existing methods fail to simultaneously achieve real-time performance, small memory footprint, and high-quality rendering for challenging real-world scenes. To address these issues, we present HyperReel -- a novel 6-DoF video representation. The two core components of HyperReel are: (1) a ray-conditioned sample prediction network that enables high-fidelity, high frame rate rendering at high resolutions and (2) a compact and memory-efficient dynamic volume representation. Our 6-DoF video pipeline achieves the best performance compared to prior and contemporary approaches in terms of visual quality with small memory requirements, while also rendering at up to 18 frames-per-second at megapixel resolution without any custom CUDA code.

PDF Abstract CVPR 2023 PDF CVPR 2023 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Novel View Synthesis DONeRF: Evaluation Dataset HyperReel PSNR 35.1 # 1
Novel View Synthesis DONeRF: Evaluation Dataset TermiNeRF PSNR 29.8 # 6
Novel View Synthesis DONeRF: Evaluation Dataset AdaNeRF PSNR 30.9 # 3
Novel View Synthesis DONeRF: Evaluation Dataset DoNeRF PSNR 30.8 # 5
Novel View Synthesis DONeRF: Evaluation Dataset Instant NGP PSNR 33.1 # 2
Novel View Synthesis DONeRF: Evaluation Dataset NeRF PSNR 30.9 # 3
Novel View Synthesis LLFF HyperReel PSNR 26.2 # 9
Novel View Synthesis LLFF TermiNeRF PSNR 23.6 # 14
Novel View Synthesis LLFF AdaNeRF PSNR 25.7 # 11
Novel View Synthesis LLFF DoNeRF PSNR 22.9 # 15
Novel View Synthesis LLFF Instant NGP PSNR 25.6 # 13

Methods