Self-Supervised Learning
Colorization
Introduced by Zhang et al. in Colorful Image ColorizationColorization is a self-supervision approach that relies on colorization as the pretext task in order to learn image representations.
Source: Colorful Image Colorization
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Colorization | 181 | 48.01% |
| Super-Resolution | 10 | 2.65% |
| Translation | 9 | 2.39% |
| Semantic correspondence | 9 | 2.39% |
| Semantic Segmentation | 9 | 2.39% |
| Image Generation | 7 | 1.86% |
| Denoising | 7 | 1.86% |
| Object Detection | 6 | 1.59% |
| Line Art Colorization | 6 | 1.59% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
