Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound (https://arxiv.org/abs/2006.11239).
Source: Denoising Diffusion Probabilistic ModelsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Generation | 94 | 13.70% |
Denoising | 81 | 11.81% |
Text-to-Image Generation | 24 | 3.50% |
Video Generation | 17 | 2.48% |
Language Modelling | 13 | 1.90% |
Super-Resolution | 12 | 1.75% |
Decoder | 12 | 1.75% |
Large Language Model | 10 | 1.46% |
3D Generation | 10 | 1.46% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |