Research map / Video / Video segmentation and tracking
NeRV: Neural Representations for Videos
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove positional encoding of frame index (None vs PE) (ablation) | NeRV · UVG ablation | PSNR: 24.93 -> 37.26 | Helped |
| PixelShuffle upscale instead of bilinear pooling (ablation) | NeRV · UVG ablation | PSNR: 29.56 -> 37.26 | Helped |
| PixelShuffle instead of transpose conv upscale (ablation) | NeRV · UVG ablation | PSNR: 36.63 -> 37.26 | Helped |
| add BatchNorm to NeRV block (ablation) | NeRV · UVG ablation | PSNR: 37.26 -> 36.71 | Hurt |
| add InstanceNorm to NeRV block (ablation) | NeRV · UVG ablation | PSNR: 37.26 -> 35.5 | Hurt |
| GELU instead of ReLU (ablation) | NeRV · UVG ablation | PSNR: 35.89 -> 37.26 | Helped |
| GELU instead of Leaky ReLU / Swish (ablation) | NeRV · UVG ablation | PSNR: 36.76 -> 37.26 | Helped |
| L1+SSIM loss instead of L2 loss (ablation) | NeRV · UVG ablation | PSNR: 35.64 -> 37.26 | Helped |
| SSIM alone vs L1+SSIM (ablation) | NeRV · UVG ablation | MS-SSIM: 0.971 vs 0.970 | Mixed |
| train NeRV-S longer, 300 to 2.4k epochs | NeRV-S · Big Buck Bunny | PSNR: 32.21 -> 34.86 | Helped |
| train NeRV-L longer, 300 to 2.4k epochs | NeRV-L · Big Buck Bunny | PSNR: 39.75 -> 41.99 | Helped |
| NeRV vs SIREN at 12.7M params | NeRV-L · Big Buck Bunny | PSNR: 25.06 -> 41.29 | Helped |
Closest papers by meaning
- DS-NeRV: Implicit Neural Video Representation with Decomposed Static and Dynamic Codes (CVPR 2024 · similarity 0.92)
- Combining Frame and GOP Embeddings for Neural Video Representation (CVPR 2024 · similarity 0.92)
- NVRC: Neural Video Representation Compression (NeurIPS 2024 · similarity 0.90)
- MetaNeRV: Meta Neural Representations for Videos with Spatial-Temporal Guidance (AAAI 2025 · similarity 0.90)
- Boosting Neural Representations for Videos with a Conditional Decoder (CVPR 2024 · similarity 0.89)
- Towards Practical Real-Time Neural Video Compression (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.