milad1378yz/MOTFM
Flow Matching for Medical Image Synthesis: Bridging the Gap Between Speed and Quality
Implements optimal transport-based flow matching with PyTorch Lightning and MONAI generative models, enabling both 2D/3D synthesis and dual mask/class conditioning. The framework uses configurable ODE solvers with adjustable inference steps and includes built-in 3D evaluation metrics (MMD, MS-SSIM, 3D-FID). Data must be prepared as a single pickled dictionary with image/mask/class splits, with inference outputs supporting multiple normalization strategies and checkpoint resolution patterns.
Stars
64
Forks
8
Language
Python
License
MIT
Category
Last pushed
Mar 16, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/milad1378yz/MOTFM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
X-GenGroup/Flow-Factory
A unified framework for easy reinforcement learning in Flow-Matching models
OpenImagingLab/FlashVSR
[CVPR 2026] Towards Real-Time Diffusion-Based Streaming Video Super-Resolution — An efficient...
fallenshock/FlowEdit
Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using...
haidog-yaqub/MeanFlow
Pytorch Implementation (unofficial) of the paper "Mean Flows for One-step Generative Modeling"...
jy0205/Pyramid-Flow
[ICLR 2025] Pyramidal Flow Matching for Efficient Video Generative Modeling