Research map / Video / Video segmentation and tracking
Reference-Based Video Super-Resolution Using Multi-Camera Video Triplets
Junyong Lee, Myeonghee Lee, Sunghyun Cho, Seungyong Lee
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 |
|---|---|---|---|
| Add multi-Ref fidelity loss with temporal window size k=7 (ablation) | RefVSR baseline · RealMCVSR test set | PSNR: 30.71 -> 31.31 | Improved |
| Add multi-Ref fidelity loss with temporal window size k=7 (ablation) | RefVSR baseline · RealMCVSR test set | SSIM: 0.894 -> 0.913 | Improved |
| Add propagative temporal fusion (PTF), including propagated matching confidence (ablation) | RefVSR with multi-Ref fidelity loss · RealMCVSR test set | PSNR: 31.31 -> 31.68 | Improved |
| Add propagative temporal fusion (PTF), including propagated matching confidence (ablation) | RefVSR with multi-Ref fidelity loss · RealMCVSR test set | SSIM: 0.913 -> 0.914 | Improved |
| Fine-tune using the adaptation loss and real-world triplet supervision (ablation) | Pre-trained RefVSR · Real-world HD video 4× super-resolution | visual fidelity | Improved |
| Add a telephoto video as an additional reference | RefVSR using a wide-angle reference · Real-world 4× video super-resolution | super-resolution quality | No clear effect |
| Use ℓ1 loss instead of the proposed perceptual-based training loss | RefVSR · RealMCVSR test set | PSNR: 31.68 -> 34.74 | Improved |
| Add information-refill and coupled propagation modules to the ℓ1-trained RefVSR | Ours-ℓ1 · RealMCVSR test set | PSNR: 34.74 -> 34.86 | Improved |
Closest papers by meaning
- Self-Supervised Learning for Real-World Super-Resolution From Dual and Multiple Zoomed Observations (TPAMI 2024 · similarity 0.89)
- Dual-Camera Super-Resolution With Aligned Attention Modules (ICCV 2021 · similarity 0.88)
- BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (CVPR 2021 · similarity 0.88)
- Frame and Feature-Context Video Super-Resolution (AAAI 2019 · similarity 0.87)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.86)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 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.
