Research map / Video / Video segmentation and tracking
Learning Spatiotemporal Frequency-Transformer for Compressed Video Super-Resolution
Zhongwei Qiu, Huan Yang, Jianlong Fu, Dongmei Fu
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 |
|---|---|---|---|
| transform frames to DCT frequency domain (Pixel+CNN to Frequency+CNN) (ablation) | BasicVSR · REDS4 compressed CRF25 | PSNR: 25.93 -> 26.87 | Improved |
| replace CNN with transformer in frequency domain (ablation) | Frequency+CNN · REDS4 compressed CRF25 | PSNR: 26.87 -> 27.15 | Improved |
| add frequency attention (FTVSR) over base transformer (ablation) | Frequency+Transformer · REDS4 compressed CRF25 | PSNR: 27.15 -> 27.28 | Improved |
| use divided time-space-frequency attention Lambda_ST vs base (ablation) | Frequency Transformer · REDS compressed CRF35 | PSNR: 24.03 -> 24.22 | Improved |
| use temporal-only frequency attention vs base (ablation) | Frequency Transformer · REDS compressed CRF25 | PSNR: 27.15 -> 27.11 | Hurt |
| use space-frequency attention Lambda_S vs base (ablation) | Frequency Transformer · REDS compressed CRF25 | PSNR: 27.15 -> 27.23 | Improved |
| vary DCT block size from 4x4 to 16x16 (ablation) | FTVSR Lambda_S · REDS compressed CRF25 | PSNR: 27.20 -> 27.20 | No clear effect |
| use FTVSR instead of COMISR | compressed VSR · REDS4 compressed CRF25 | PSNR: 26.47 -> 28.05 | Improved |
| use FTVSR instead of IconVSR | VSR · REDS4 uncompressed | PSNR: 31.67 -> 31.82 | Improved |
| train FTVSR on both compressed and uncompressed videos | FTVSR · REDS4 uncompressed | PSNR: 31.82 -> 31.74 | Hurt |
| train FTVSR on both compressed and uncompressed videos | FTVSR · Vid4 uncompressed | PSNR: 28.31 -> 28.06 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- COMISR — COMISR: Compression-Informed Video Super-Resolution (ICCV 2021) · better on 5 of 7 shared results
- BasicVSR — BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (CVPR 2021) · better on 3 of 3 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
- Learning Degradation-Robust Spatiotemporal Frequency-Transformer for Video Super-Resolution (TPAMI 2023 · similarity 0.96)
- Learning Trajectory-Aware Transformer for Video Super-Resolution (CVPR 2022 · similarity 0.89)
- Towards Interpretable Video Super-Resolution via Alternating Optimization (ECCV 2022 · similarity 0.88)
- Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution (CVPR 2026 · similarity 0.88)
- RealViformer: Investigating Attention for Real-World Video Super-Resolution (ECCV 2024 · similarity 0.87)
- Turning Frequency to Resolution: Video Super-Resolution via Event Cameras (CVPR 2021 · similarity 0.87)
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.
