Research map / Video / Video segmentation and tracking
LDIP: Long Distance Information Propagation for Video Super-Resolution
Michael Bernasconi, Abdelaziz Djelouah, Yang Zhang, Markus Gross, Christopher Schroers
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 LRRF module trained with HQ references (vs HRIP only) (ablation) | BasicVSR++ backbone · REDS x4 ablation | LPIPS: 0.129 -> 0.106 | Improved |
| add LRRF module trained with HQ references (vs HRIP only) (ablation) | BasicVSR++ backbone · REDS x4 ablation | PSNR: 32.28 -> 32.04 | Hurt |
| use HQ references instead of self-referential references (ablation) | BasicVSR++ backbone · REDS x4 ablation | LPIPS: 0.126 -> 0.106 | Improved |
| add self-referential references to HQ training (ablation) | BasicVSR++ backbone · REDS x4 ablation | LPIPS: 0.106 -> 0.126 | Hurt |
| use RAFT instead of SpyNet for HRIP flow (ablation) | BasicVSR++ backbone · REDS x4 ablation | LPIPS: 0.110 -> 0.107 | Improved |
| use PDCNetPlus instead of RAFT for HRIP flow (ablation) | BasicVSR++ backbone · REDS x4 ablation | LPIPS: 0.107 -> 0.106 | No clear effect |
| replace upscaling layer with arbitrary scaling module (ablation) | BasicVSR++ · REDS x4 ablation | PSNR: 32.41 -> 32.13 | Hurt |
| add long range propagation on top of arbitrary scaling (ablation) | BasicVSR++ with AS · REDS x4 ablation | LPIPS: 0.129 -> 0.113 | Improved |
| add long range propagation on top of arbitrary scaling (ablation) | BasicVSR++ with AS · REDS x4 ablation | PSNR: 32.13 -> 31.67 | Hurt |
| add LDIP to IART backbone | IART · REDS x4 | LPIPS: 0.105 -> 0.097 | Improved |
| add LDIP to IART backbone | IART · REDS x4 | PSNR: 33.20 -> 32.83 | Hurt |
| 5 references every 5 frames instead of every 1 (ablation) | Ours (IART) · REDS x4 key-frame density | memory (GB): 31.1 -> 22.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- BasicVSR++ — BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (CVPR 2021) · better on 4 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PatchVSR: Breaking Video Diffusion Resolution Limits with Patch-wise Video Super-Resolution (CVPR 2025 · similarity 0.87)
- SeeClear: Semantic Distillation Enhances Pixel Condensation for Video Super-Resolution (NeurIPS 2024 · similarity 0.87)
- VideoGigaGAN: Towards Detail-rich Video Super-Resolution (CVPR 2025 · similarity 0.87)
- Arbitrary-Scale Video Super-resolution Guided by Dynamic Context (AAAI 2024 · similarity 0.87)
- EvSTVSR: Event Guided Space-Time Video Super-Resolution (AAAI 2025 · similarity 0.87)
- One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-Resolution (NeurIPS 2025 · similarity 0.86)
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.
