Spaces:
Running
Download README.md from JohannesK14/RIPE: direct link, hf CLI and curl.
- Browser
- Download file 2.78 kB
-
https://huggingface.co/spaces/JohannesK14/RIPE/resolve/main/README.md
- Command line
-
hf download hf://spaces/JohannesK14/RIPE/README.md
-
curl -L -o README.md https://huggingface.co/spaces/JohannesK14/RIPE/resolve/main/README.md
A newer version of the Gradio SDK is available: 6.32.0
title: RIPE & RIPE++ Keypoint Matching
emoji: 📚
colorFrom: gray
colorTo: indigo
sdk: gradio
sdk_version: 5.35.0
app_file: app.py
pinned: false
license: other
short_description: Demo of RIPE (ICCV 2025) and RIPE++ (LIMIT @ ECCV 2026).
RIPE & RIPE++ Keypoint Matching Demo
A Gradio demo for two keypoint extractors from our group, side by side:
- RIPE — Reinforcement Learning on Unlabeled Image Pairs for Robust Keypoint Extraction (ICCV 2025). arXiv · Project page · Code
- RIPE++ — our latest model (LIMIT @ ECCV 2026, Malmö). arXiv · Project page · Code
Pick a method, upload an image pair (or choose an example), and the demo:
- extracts the top-2048 keypoints from each image,
- matches them with the selected matcher,
- filters outliers with RANSAC (fundamental matrix) when more than 8 matches are found,
- visualizes the detected keypoints and the raw / RANSAC-filtered correspondences.
Images are resized so the smaller side lies within [512, 768] px, aspect ratio preserved.
Matchers
- Nearest Neighbor — mutual nearest-neighbor matching (kornia MNN for RIPE; the glue-factory NN matcher for RIPE++).
- LightGlue — our trained LightGlue matcher (from glue-factory). Available for RIPE++ only.
Both models are trained without geometric ground truth (no depth, pose, or annotated correspondences) — only image pairs and a geometric reward.
How it's built
The method source is vendored directly in this repo — ripe/ (RIPE) and ripepp/ +
gluefactory/ (RIPE++) — so nothing is installed from git at build time. Model weights
are vendored too:
weights_ripe.pth— RIPE.weights/ripe++.ckpt— RIPE++ extractor.weights/ripe++lightglue.tar— RIPE++ LightGlue matcher.
The app is decorated with @spaces.GPU, so it runs on ZeroGPU: the GPU is attached only
for the duration of each inference call.
Licenses
ripe/andweights_ripe.pth— RIPE, Fraunhofer academic-use license (see the repo-rootLICENSE).ripepp/(+ RIPE++ weights) — RIPE++, Fraunhofer academic-use license (ripepp/LICENSE).gluefactory/— vendored from the glue-factory fork (based on cvg/glue-factory), Apache License 2.0 (gluefactory/LICENSE).
Everything else is installed from PyPI (see requirements.txt) under its own license.
Running locally
pip install -r requirements.txt
python app.py