mhamilton723/STEGO
Unsupervised Semantic Segmentation by Distilling Feature Correspondences
This project helps you automatically identify and label different objects or regions within images, even if you don't have existing labels for what's in them. You provide a collection of images, and it outputs images where each pixel is assigned to a specific category, like 'sky,' 'car,' or 'road.' This is ideal for researchers, scientists, or analysts working with large image datasets where manual labeling is time-consuming or impractical.
785 stars. No commits in the last 6 months.
Use this if you need to automatically segment objects within images without the laborious and often impossible task of manually labeling every pixel in your dataset.
Not ideal if you already have perfectly labeled datasets or require pixel-perfect segmentation accuracy that only extensive human annotation can provide.
Stars
785
Forks
159
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 24, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mhamilton723/STEGO"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
deepinv/deepinv
DeepInverse: a PyTorch library for solving imaging inverse problems using deep learning
fidler-lab/polyrnn-pp
Inference Code for Polygon-RNN++ (CVPR 2018)
yjxiong/tsn-pytorch
Temporal Segment Networks (TSN) in PyTorch
pyxu-org/pyxu
Modular and scalable computational imaging in Python with GPU/out-of-core computing.
oeway/pytorch-deform-conv
PyTorch implementation of Deformable Convolution