microsoft/ProDA
Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation (CVPR 2021)
This project helps computer vision engineers apply semantic segmentation models trained on synthetic datasets (like GTA5 or SYNTHIA) to real-world images (like Cityscapes). It takes an existing segmentation model and a collection of real-world images, then produces a more accurate version of the model that performs better on the real-world data. This is for machine learning practitioners and researchers focused on deploying image segmentation in varied environments.
291 stars. No commits in the last 6 months.
Use this if you need to adapt a semantic segmentation model trained on one type of image (source domain) to accurately segment objects in a different type of real-world image (target domain) without extensive manual labeling of the target images.
Not ideal if you are working with non-image data, do not have access to large synthetic datasets, or need a segmentation model for a domain where no similar source data exists.
Stars
291
Forks
44
Language
Python
License
MIT
Category
Last pushed
Jul 06, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/microsoft/ProDA"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
adapt-python/adapt
Awesome Domain Adaptation Python Toolbox
corenel/pytorch-adda
A PyTorch implementation for Adversarial Discriminative Domain Adaptation
jindongwang/transferlearning
Transfer learning / domain adaptation / domain generalization / multi-task learning etc. Papers,...
thuml/Transfer-Learning-Library
Transfer Learning Library for Domain Adaptation, Task Adaptation, and Domain Generalization
KaiyangZhou/Dassl.pytorch
A PyTorch toolbox for domain generalization, domain adaptation and semi-supervised learning.