Research map / Video / Video segmentation and tracking
Perceiving Better Moments: Cover Frame Reselection and Enhancement for Live Photos with the Live2K Dataset
Junyu Lou, Kai Chen, Weiyi You, Hui Zeng, Lei Zhang, Shuhang Gu
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 multi-frame input (9 frames with TSA fusion) (ablation) | LPENet · Live2K OPPO | PSNR: 27.13 -> 27.42 | Improved |
| add intermediate color supervision loss (ablation) | LPENet · Live2K OPPO | PSNR: 27.42 -> 27.48 | Improved |
| add the cover reference image with guided enhancement (ablation) | LPENet · Live2K OPPO | PSNR: 27.48 -> 27.78 | Improved |
| add color modulation alone to guided enhancement (ablation) | LPENet · Live2K OPPO | PSNR: 27.48 -> 27.66 | Improved |
| add texture transfer alone to guided enhancement (ablation) | LPENet · Live2K OPPO | PSNR: 27.48 -> 27.68 | Improved |
| combine color modulation and texture transfer (ablation) | LPENet · Live2K OPPO | PSNR: 27.48 -> 27.78 | Improved |
| LPENet versus strongest baselines trained on Live2K | LPENet · Live2K iPhone | PSNR: 26.58 -> 26.97 | Improved |
| burst SR methods versus single-image SwinIR | BIPNet/Burstormer/BurstM · Live2K iPhone | PSNR: 26.58 -> 25.97 | Hurt |
| fine-tune OPPO-pretrained model on 60 iPhone images for 5,000 iterations | LPENet · Live2K iPhone | PSNR | Improved |
Closest papers by meaning
- LiveMoments: Reselected Key Photo Restoration in Live Photos via Reference-guided Diffusion (ICLR 2026 · similarity 0.89)
- Learning Video Exposure Correction: Large-Scale Real-World Dataset and Method (TPAMI 2026 · similarity 0.85)
- Learning Spatial Adaptation and Temporal Coherence in Diffusion Models for Video Super-Resolution (CVPR 2024 · similarity 0.83)
- Removing Reflections from RAW Photos (CVPR 2025 · similarity 0.83)
- VideoElevator: Elevating Video Generation Quality with Versatile Text-to-Image Diffusion Models (AAAI 2025 · similarity 0.82)
- Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution (CVPR 2024 · similarity 0.82)
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.
