Research map / Video / Video segmentation and tracking
Fully Automatic Video Colorization With Self-Regularization and Diversity
Chenyang Lei, Qifeng Chen
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 self-regularization instead of removing it (ablation) | Ours · DAVIS | Preference rate: 67.9% | Improved |
| Add diversity loss (ablation) | Ours · DAVIS | Preference rate: 61.5% | Improved |
| Apply BTC temporal consistency refinement | Zhang et al. · DAVIS and Videvo | LPIPS and PSNR: DAVIS LPIPS 0.218 -> 0.243; PSNR 29.25 -> 29.07; Videvo LPIPS 0.201 -> 0.249; PSNR 29.52 -> 29.04 | Hurt |
| Apply BTC temporal consistency refinement | Iizuka et al. · DAVIS and Videvo | LPIPS and PSNR: DAVIS LPIPS 0.189 -> 0.218; PSNR 29.91 -> 29.25; Videvo LPIPS 0.190 -> 0.241; PSNR 30.23 -> 28.90 | Hurt |
| Use the proposed video colorization method rather than Zhang et al. + BTC | Fully automatic video colorization · DAVIS and Videvo | Perceptual preference rate: 80.0%; 88.8% | Improved |
| Use the proposed video colorization method rather than Iizuka et al. + BTC | Fully automatic video colorization · DAVIS and Videvo | Perceptual preference rate: 72.8%; 63.3% | Improved |
Closest papers by meaning
- Deep Exemplar-Based Video Colorization (CVPR 2019 · similarity 0.89)
- Learning Representations for Automatic Colorization (ECCV 2016 · similarity 0.88)
- Learned Variational Video Color Propagation (ECCV 2022 · similarity 0.88)
- Colorful Image Colorization (ECCV 2016 · similarity 0.87)
- ColorMNet: A Memory-based Deep Spatial-Temporal Feature Propagation Network for Video Colorization (ECCV 2024 · similarity 0.87)
- Colorization Transformer (ICLR 2021 · similarity 0.87)
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.
