This repository contains the code for the paper "RayEmb: Arbitrary Landmark Detection in X-Ray Images Using Ray Embedding Subspace," accepted as an oral presentation at ACCV 2024.
RayEmb introduces a novel approach for detecting arbitrary landmarks in X-ray images using ray embedding subspace. Our approach represents 3D points as distinct subspaces, formed by feature vectors (referred to as ray embeddings) corresponding to intersecting rays. Establishing 2D-3D correspondences then becomes a task of finding ray embeddings that are close to a given subspace, essentially performing an intersection test.
- Synthetic image generation
- Rayemb model training and inference
- CLI based usage
The code and documentation are currently being prepared for release. Please stay tuned for updates.
- Installation instructions
- Example usage
- Pretrained models
- Visualization scripts
For any questions or collaboration inquiries, please contact shrestha.pragyan@image.iit.tsukuba.ac.jp.
Stay tuned for more updates!