This is a reimplementation of the paper Sparse Subspace Clustering: Algorithm, Theory, and Applications with CUDA support.
Some of the implementation is based on the Matlab SSC-Basis Pursuit ADMM code from the author's website.
The SSC implementation in ssc.ssc.sparse_subspace_clustering.
The corresponding ADMM implementation in ssc/admm.py(CPU) and ssc/admm_gpu.py(CUDA).
The experiment from the original paper is in face_clustering.py and motion_segmenation.py
Please install follow the instructions in CuPY install guide.
According to your CUDA version, you will use pip install cupy-cuda12x or pip install cupy-cuda11x.
Other requirements are in requirements.txt.
The following datasets should be placed in the ./data folder to reproduce the experiments.
Hopkins155YaleBCrop025.matis in the original Matlab code.