Generative Models
Conditional / Rectified flow matching
Introduced by Lipman et al. in Flow Matching for Generative ModelingConditional Flow Matching (CFM) is a fast way to train continuous normalizing flow (CNF) models. CFM is a simulation-free training objective for continuous normalizing flows that allows conditional generative modelling and speeds up training and inference.
Source: Flow Matching for Generative Modeling
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Generation | 6 | 14.29% |
| Denoising | 4 | 9.52% |
| Speech Synthesis | 4 | 9.52% |
| Text-To-Speech Synthesis | 4 | 9.52% |
| Conditional Image Generation | 2 | 4.76% |
| Depth Estimation | 2 | 4.76% |
| Protein Design | 2 | 4.76% |
| Optical Flow Estimation | 2 | 4.76% |
| Image-to-Image Translation | 1 | 2.38% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
