Research map / Video / Video segmentation and tracking
Onion-Peel Networks for Deep Video Completion
Seoung Wug Oh, Sungho Lee, Joon-Young Lee, Seon Joo Kim
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 |
|---|---|---|---|
| fill the hole progressively by onion peeling instead of one-shot (ablation) | Onion-Peel Network · video completion | output quality and speed: one-shot variant runs about 3x faster than onion-peel network (p=8) but quality is worse | Improved |
| add temporal consistency post-processing (ablation) | Onion-Peel Network · DAVIS video completion | temporal coherence | Mixed |
| use asymmetric attention over reference frames instead of optical flow | Onion-Peel Network · DAVIS video completion | PSNR: 23.34 -> 30.19 vs Yu et al. and 28.04 -> 30.19 vs Huang et al. | Improved |
| compare against global-flow optimization method in user study | Onion-Peel Network · DAVIS user study | average rank: 1.93 vs 1.95 | No clear effect |
Closest papers by meaning
- Copy-and-Paste Networks for Deep Video Inpainting (ICCV 2019 · similarity 0.85)
- Deep Flow-Guided Video Inpainting (CVPR 2019 · similarity 0.84)
- Completion as Enhancement: A Degradation-Aware Selective Image Guided Network for Depth Completion (CVPR 2025 · similarity 0.84)
- Deep Video Inpainting (CVPR 2019 · similarity 0.84)
- Contextual-based Image Inpainting: Infer, Match, and Translate (ECCV 2018 · similarity 0.83)
- Efficient Indoor Depth Completion Network Using Mask-adaptive Gated Convolution (AAAI 2025 · 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.
