Research map / Video / Video segmentation and tracking
Video Rescaling Networks With Joint Optimization Strategies for Downscaling and Upscaling
Yan-Cheng Huang, Yi-Hsin Chen, Cheng-You Lu, Hui-Po Wang, Wen-Hsiao Peng, Ching-Chun Huang
Check 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 |
|---|---|---|---|
| jointly optimize downscaling and upscaling rather than use predefined downscaling | LSTM-VRN versus video super-resolution baselines · Vid4 | PSNR-Y: 27.33-27.92dB -> 32.24dB | Improved |
| use LR temporal context to predict missing high-frequency components | LSTM-VRN versus IRN · Vid4 | PSNR-Y: 31.29 -> 32.24; 0.95dB | Improved |
| use GoF-based multi-input, multi-output processing for both downscaling and upscaling | MIMO-VRN versus LSTM-VRN · Vid4 | PSNR-Y: 32.24 -> 33.79; 1.55dB | Improved |
| use GoF-based multi-input, multi-output processing for both downscaling and upscaling | MIMO-VRN versus LSTM-VRN · Vimeo-90K-T | PSNR-Y: 41.42 -> 43.26; 1.84dB | Improved |
| reset recurrent propagation within a sliding window (ablation) | LSTM-VRN · Vid4 | PSNR-Y: 31.16 -> 31.53 | Improved |
| add backward propagation to forward-only sliding-window propagation (ablation) | LSTM-VRN · Vid4 | PSNR-Y: 31.53 -> 32.24 | Improved |
| increase GoF size from 1 to 3 (ablation) | MIMO-VRN · Vid4 | PSNR-Y: 30.69 -> 33.61 | Improved |
| increase GoF size from 3 to 5 (ablation) | MIMO-VRN · Vid4 | PSNR-Y: 33.61 -> 33.79 | Improved |
| increase GoF size from 5 to 7 (ablation) | MIMO-VRN · Vid4 | PSNR-Y: 33.79 -> 33.45 | Hurt |
| add center loss on HR reconstruction (ablation) | MIMO-VRN · Vid4 | PSNR-Y: 33.79 -> 33.40 | Hurt |
| add center loss on HR reconstruction (ablation) | MIMO-VRN · Vid4 | MAD (HR): 0.88 -> 0.28 | Improved |
| add center loss on LR video (ablation) | MIMO-VRN · Vid4 | PSNR-Y: 45.05 -> 42.42 | Hurt |
Closest papers by meaning
- Downscaled Representation Matters: Improving Image Rescaling with Collaborative Downscaled Images (ICCV 2023 · similarity 0.88)
- Learning Spatio-Temporal Downsampling for Effective Video Upscaling (ECCV 2022 · similarity 0.87)
- Invertible Image Rescaling (ECCV 2020 · similarity 0.87)
- Plug-and-Play Tri-Branch Invertible Block for Image Rescaling (AAAI 2025 · similarity 0.85)
- Deep Generative Model based Rate-Distortion for Image Downscaling Assessment (CVPR 2024 · similarity 0.84)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · 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.
