Research map / Video / Video segmentation and tracking
Inertia-Guided Flow Completion and Style Fusion for 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 |
|---|---|---|---|
| Use inertia-guided feature-domain flow alignment and aggregation (ablation) | IGFC · DAVIS square masks | EPE: 0.58 -> 0.56 | Helped |
| Use inertia-guided feature-domain flow alignment and aggregation (ablation) | IGFC · DAVIS object masks | PSNR: 34.90 -> 35.16 | Helped |
| Apply inertia warping in the feature domain rather than the flow domain (ablation) | IGFC · DAVIS square masks | PSNR: 32.91 -> 33.23 | Helped |
| Apply inertia warping in the feature domain rather than the flow domain (ablation) | IGFC · DAVIS object masks | EPE: 0.38 -> 0.35 | Helped |
| Add ternary census transform loss to flow completion training (ablation) | IGFC · DAVIS | flow motion-boundary clarity | Helped |
| Add ASFN to refine warped gradients using style statistics from valid regions (ablation) | Ours · DAVIS square masks | PSNR: 32.91 -> 33.23 | Helped |
| Add ASFN to refine warped gradients using style statistics from valid regions (ablation) | Ours · DAVIS object masks | PSNR: 34.88 -> 35.16 | Helped |
| Add ASFN to the DFGVI flow-guided framework (ablation) | DFGVI · DAVIS square masks | PSNR: 31.24 -> 31.22 | No clear effect |
| Add ASFN to the DFGVI flow-guided framework (ablation) | DFGVI · DAVIS object masks | PSNR: 33.12 -> 33.23 | Helped |
| Add ASFN to the FGVC flow-guided framework (ablation) | FGVC · DAVIS square masks | PSNR: 32.14 -> 32.37 | Helped |
| Add ASFN to the FGVC flow-guided framework (ablation) | FGVC · DAVIS object masks | PSNR: 33.91 -> 34.17 | Helped |
Closest papers by meaning
- Video Diffusion Models Are Strong Video Inpainter (AAAI 2025 · similarity 0.86)
- FlowVid: Taming Imperfect Optical Flows for Consistent Video-to-Video Synthesis (CVPR 2024 · similarity 0.85)
- Injecting Frame-Event Complementary Fusion into Diffusion for Optical Flow in Challenging Scenes (NeurIPS 2025 · similarity 0.85)
- Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 2024 · similarity 0.84)
- Disentangled Motion Modeling for Video Frame Interpolation (AAAI 2025 · similarity 0.84)
- Bridge Frame and Event: Common Spatiotemporal Fusion for High-Dynamic Scene Optical Flow (CVPR 2025 · 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.