Research map / Video / Video segmentation and tracking
Bringing Old Films Back to Life
Check 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 |
|---|---|---|---|
| use RTN instead of the strongest baseline, BasicVSR | Synthetic old-film restoration · Synthetic restoration dataset (DAVIS) | PSNR: 23.363 -> 24.465 | Helped |
| use RTN instead of the strongest baseline, BasicVSR | Synthetic old-film restoration · Synthetic restoration dataset (DAVIS) | LPIPS: 0.328 -> 0.192 | Helped |
| use RTN instead of the strongest baseline, Old Photo+TS | Synthetic old-film restoration · Synthetic restoration dataset (DAVIS) | E_warp: 0.0041 -> 0.0019 | Helped |
| use RTN instead of the strongest baseline, Old Photo+TS | Real old-film restoration · Real old films | NIQE: 17.5110 -> 15.4254 | Helped |
| use RTN instead of the strongest baseline, Old Photo+TS | Real old-film restoration · Real old films | BRISQUE: 48.1470 -> 42.1422 | Helped |
| use RTN instead of DeepExemplar | Reference-based video colorization · REDS colorization | FID: 37.971 -> 31.992 | Helped |
| add bidirectional recurrence using past and future frames (ablation) | RTN · Synthetic restoration dataset (DAVIS) | E_warp: 0.0036 -> 0.0019 | Helped |
| add the learned soft mask instead of channel-wise concatenation (ablation) | RTN · Synthetic restoration dataset (DAVIS) | SSIM: 0.827 -> 0.840 | Helped |
| add transformer layers instead of CNN-based spatial restoration (ablation) | RTN · Synthetic restoration dataset (DAVIS) | LPIPS: 0.229 -> 0.192 | Helped |
Closest papers by meaning
- Making Old Film Great Again: Degradation-aware State Space Model for Old Film Restoration (CVPR 2025 · similarity 0.88)
- Learning Truncated Causal History Model for Video Restoration (NeurIPS 2024 · similarity 0.84)
- Direction-Aware Video Demoiréing with Temporal-Guided Bilateral Learning (AAAI 2024 · similarity 0.84)
- ReFIR: Grounding Large Restoration Models with Retrieval Augmentation (NeurIPS 2024 · similarity 0.84)
- SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration (CVPR 2025 · similarity 0.84)
- Memory Consolidation Enables Long-Context Video Understanding (ICML 2024 · similarity 0.83)
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.