Research map / Video / Video segmentation and tracking
DLFormer: Discrete Latent Transformer 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 discrete latent code inference instead of continuous-feature video inpainting (ablation) | DLFormer · Youtube-VOS | PSNR: 33.07 -> 33.95 | Helped |
| use discrete latent code inference instead of continuous-feature video inpainting (ablation) | DLFormer · Youtube-VOS | VFID: 0.124 -> 0.082 | Helped |
| add the temporal aggregation block (TAB) (ablation) | DLFormer · Youtube-VOS | PSNR: 33.82 -> 33.95 | Helped |
| add the temporal aggregation block (TAB) (ablation) | DLFormer · Youtube-VOS | SSIM: 0.968 -> 0.970 | Helped |
| add the temporal aggregation block (TAB) (ablation) | DLFormer · Youtube-VOS | VFID: 0.086 -> 0.082 | Helped |
| use DLFormer rather than STTN | video restoration · Youtube-VOS | PSNR: 33.67 -> 33.95 | Helped |
| use DLFormer rather than STTN | video restoration · Youtube-VOS | VFID: 0.087 -> 0.082 | Helped |
| use DLFormer rather than FuseFormer | video restoration · DAVIS | PSNR: 33.45 -> 34.22 | Helped |
| use DLFormer rather than FuseFormer | video restoration · DAVIS | SSIM: 0.979 -> 0.977 | Hurt |
| use DLFormer rather than STTN | video restoration · DAVIS | VFID: 0.071 -> 0.062 | Helped |
Closest papers by meaning
- WaveFormer: Wavelet Transformer for Noise-Robust Video Inpainting (AAAI 2024 · similarity 0.89)
- Don't Look into the Dark: Latent Codes for Pluralistic Image Inpainting (CVPR 2024 · similarity 0.87)
- Video Diffusion Models Are Strong Video Inpainter (AAAI 2025 · similarity 0.87)
- Reconstruct, Inpaint, Test-Time Finetune: Dynamic Novel-view Synthesis from Monocular Videos (NeurIPS 2025 · similarity 0.86)
- SyFormer: Structure-Guided Synergism Transformer for Large-Portion Image Inpainting (AAAI 2024 · similarity 0.86)
- MotionAura: Generating High-Quality and Motion Consistent Videos using Discrete Diffusion (ICLR 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.