Research map / Video / Video segmentation and tracking
Deep Stereo Video Inpainting
Read the paperCheck 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 SVINet instead of E2FGVI | stereo video inpainting · KITTI2012 | PSNR: 29.3312 -> 29.6236 | Helped |
| Use SVINet instead of E2FGVI | stereo video inpainting · KITTI2015 | PSNR: 29.5729 -> 30.8191 | Helped |
| Add temporal alignment rather than using no alignment (ablation) | SVINet · KITTI2015 | PSNR: 25.7103 -> 30.8191 | Helped |
| Use flow-guided deformable alignment rather than flow warping (ablation) | SVINet · KITTI2015 | PSNR: 29.4325 -> 30.2413 | Helped |
| Use flow-guided deformable alignment rather than DCN alone (ablation) | SVINet · KITTI2015 | PSNR: 30.1276 -> 30.2413 | Helped |
| Use lightweight motion-estimator flow rather than externally generated flow to guide DCN (ablation) | SVINet · KITTI2015 | PSNR: 30.2413 -> 30.2206 | No clear effect |
| Aggregate aligned features with original features (ablation) | SVINet · KITTI2015 | PSNR: 30.2206 -> 30.3427 | Helped |
| Cascade temporal alignment modules (ablation) | SVINet · KITTI2015 | PSNR: 30.3427 -> 30.5876 | Helped |
| Add self-supervised alignment loss (ablation) | SVINet · KITTI2015 | PSNR: 30.5876 -> 30.8191 | Helped |
| Add stereo consistency loss (ablation) | SVINet · KITTI2015 | EPE: 0.3065 -> 0.2668 | Helped |
| Add cross-view feature aggregation (ablation) | SVINet · KITTI2015 | PSNR: 30.6875 -> 30.8191 | Helped |
Closest papers by meaning
- DreamStereo: Towards Real-Time Stereo Inpainting for HD Videos (CVPR 2026 · similarity 0.88)
- Deep Video Inpainting (CVPR 2019 · similarity 0.88)
- Video Inpainting by Jointly Learning Temporal Structure and Spatial Details (AAAI 2019 · similarity 0.87)
- Progressive Temporal Feature Alignment Network for Video Inpainting (CVPR 2021 · similarity 0.87)
- Semi-Supervised Video Inpainting With Cycle Consistency Constraints (CVPR 2023 · similarity 0.87)
- MVInpainter: Learning Multi-View Consistent Inpainting to Bridge 2D and 3D Editing (NeurIPS 2024 · 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.