Research map / Video / Video generation
GIViC: Generative Implicit Video Compression
Ge Gao, Siyue Teng, Tianhao Peng, Fan Zhang, David Bull
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add 32-frame Vimeo raw fine-tuning instead of 7-frame Vimeo-90k (ablation) | GIViC · UVG | BD-rate PSNR (%): 5.87 | Improved |
| add overfitting in the encoding pipeline (ablation) | GIViC · UVG | BD-rate PSNR (%): 3.66 | Improved |
| add temporal inflation of pretrained image autoencoder instead of from-scratch 3D autoencoder (ablation) | GIViC · UVG | BD-rate PSNR (%): 0.98 | Improved |
| add multi-resolution implicit diffusion instead of single-resolution per-token diffusion (ablation) | GIViC · UVG | BD-rate PSNR (%): 2.35 | Improved |
| add implicit diffusion instead of RelayDiffusion (ablation) | GIViC · UVG | BD-rate PSNR (%): 2.53 | Improved |
| add consistency objective instead of flow matching (ablation) | GIViC · UVG | BD-rate PSNR (%): 3.09 | Improved |
| add hierarchical lower bound in HGLA instead of original formulation (ablation) | GIViC · UVG | BD-rate PSNR (%): 1.99 | Improved |
| add learned gating across layers in HGLA (ablation) | GIViC · UVG | BD-rate PSNR (%): 3.17 | Improved |
| add HGLA instead of vanilla transformer with next-scale prediction (ablation) | GIViC · UVG | BD-rate PSNR (%): 0.19 | Improved |
| add HGLA context model instead of convolution-based context model (ablation) | GIViC · UVG | BD-rate PSNR (%): 6.77 | Improved |
| add shared context model instead of instance-specific overfitting of it (ablation) | GIViC · UVG | BD-rate PSNR (%): 3.20 | Improved |
Closest papers by meaning
- HiNeRV: Video Compression with Hierarchical Encoding-based Neural Representation (NeurIPS 2023 · similarity 0.87)
- NVRC: Neural Video Representation Compression (NeurIPS 2024 · similarity 0.86)
- PNVC: Towards Practical INR-based Video Compression (AAAI 2025 · similarity 0.85)
- Boosting Neural Representations for Videos with a Conditional Decoder (CVPR 2024 · similarity 0.84)
- Combining Frame and GOP Embeddings for Neural Video Representation (CVPR 2024 · similarity 0.84)
- Towards Practical Real-Time Neural Video Compression (CVPR 2025 · similarity 0.84)
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.
