Research map / Video / Video generation
Diversified Dynamical Gaussian Process Latent Variable Model for Video Repair
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 |
|---|---|---|---|
| Add diversity-encouraging prior over inducing points to DGPLVM (D2GPLVM) (ablation) | DGPLVM · Hollywood2 block damage | MAE: 9.53 -> 7.33 | Helped |
| Add diversity-encouraging prior over inducing points to DGPLVM (D2GPLVM) (ablation) | DGPLVM · Hollywood2 scattered damage | MAE: 10.73 -> 7.91 | Helped |
| Add diversity-encouraging prior to DGPLVM (ablation) | DGPLVM · Hollywood2 block damage | SSIM: 0.84 -> 0.88 | Helped |
| Add diversity-encouraging prior to DGPLVM (ablation) | DGPLVM · Hollywood2 scattered damage | SSIM: 0.90 -> 0.95 | Helped |
| Add dynamical sequential prior to BGPLVM (DGPLVM) (ablation) | BGPLVM · Hollywood2 block damage | MAE: 10.68 -> 9.53 | Helped |
| Add dynamical sequential prior to BGPLVM (DGPLVM) (ablation) | BGPLVM · Hollywood2 scattered damage | MAE: 11.84 -> 10.73 | Helped |
| Add dynamical sequential prior to BGPLVM (DGPLVM) (ablation) | BGPLVM · Hollywood2 block damage | SSIM: 0.83 -> 0.84 | Helped |
| Replace exemplar-based inpainting with probabilistic global temporal model D2GPLVM | EB · Hollywood2 scattered damage | SSIM: 0.37 -> 0.95 | Helped |
| Use D2GPLVM instead of exemplar-based inpainting for repair time | D2GPLVM · Hollywood2 clips | repair time: over 15 minutes -> around two seconds | Helped |
| Use D2GPLVM instead of DGPLVM/BGPLVM for repair time | D2GPLVM · Hollywood2 clips | repair time | No clear effect |
Closest papers by meaning
- Learning Efficient Fuse-and-Refine for Feed-Forward 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.83)
- Discrete Prior-Based Temporal-Coherent Content Prediction for Blind Face Video Restoration (AAAI 2025 · similarity 0.82)
- Learning Truncated Causal History Model for Video Restoration (NeurIPS 2024 · similarity 0.82)
- Noise-Resistant Video Anomaly Detection via RGB Error-Guided Multiscale Predictive Coding and Dynamic Memory (CVPR 2025 · similarity 0.82)
- 4D Gaussian Splatting in the Wild with Uncertainty-Aware Regularization (NeurIPS 2024 · similarity 0.82)
- Warped Diffusion: Solving Video Inverse Problems with Image Diffusion Models (NeurIPS 2024 · 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.