Research map / Video / Video segmentation and tracking
Towards an End-to-End Framework for Flow-Guided Video Inpainting
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace FuseFormer with the end-to-end flow-guided framework | Video inpainting · YouTube-VOS test | PSNR: 33.29 -> 33.71 | Helped |
| Replace FuseFormer with the end-to-end flow-guided framework | Video inpainting · DAVIS test | VFID: 0.138 -> 0.116 | Helped |
| Remove motion supervision by removing flow consistency loss (ablation) | E²FGVI flow completion · DAVIS | PSNR: 32.35 -> 32.08 | Hurt |
| Add flow completion to the pretrained flow estimates (ablation) | E²FGVI flow completion · DAVIS | PSNR: 32.23 -> 32.35 | Helped |
| Add flow-guided feature propagation (ablation) | E²FGVI · DAVIS | PSNR: 31.73 -> 32.15 | Helped |
| Add deformable convolution to flow-based propagation (ablation) | E²FGVI feature propagation · DAVIS | PSNR: 32.15 -> 32.35 | Helped |
| Remove flow guidance from deformable-convolution propagation (ablation) | E²FGVI feature propagation · DAVIS | PSNR: 32.35 -> 32.17 | Hurt |
| Replace focal attention with local window attention (ablation) | E²FGVI attention mechanism · DAVIS | PSNR: 31.73 -> 31.57 | Hurt |
| Replace focal attention with vanilla global attention (ablation) | E²FGVI attention mechanism · DAVIS | PSNR: 31.73 -> 31.74 | Helped |
| Replace online completion with offline FGVC-completed flows (ablation) | E²FGVI flow completion · DAVIS | PSNR: 32.35 -> 32.38 | Helped |
| Replace online completion with offline FGVC-completed flows (ablation) | E²FGVI flow completion · DAVIS | runtime (s/frame): 0.16 -> 1.21 | Hurt |
| Use the proposed hallucination module instead of an image inpainting model after flow pre-fill (ablation) | DAVIS video inpainting · DAVIS | PSNR: 30.80 -> 31.74 | Helped |
Closest papers by meaning
- Elevating Flow-Guided Video Inpainting with Reference Generation (AAAI 2025 · similarity 0.89)
- Video Diffusion Models Are Strong Video Inpainter (AAAI 2025 · similarity 0.89)
- Keyframe-Guided Creative Video Inpainting (CVPR 2025 · similarity 0.87)
- Infinite-Canvas: Higher-Resolution Video Outpainting with Extensive Content Generation (AAAI 2025 · similarity 0.86)
- FlowVid: Taming Imperfect Optical Flows for Consistent Video-to-Video Synthesis (CVPR 2024 · similarity 0.86)
- CoCoCo: Improving Text-Guided Video Inpainting for Better Consistency, Controllability and Compatibility (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.