Reproduce “Fixed-Rank Representation for Unsupervised Visual Learning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Risheng Liu, Zhouchen Lin, Fernando De la Torre, Zhixun Su
FRR finds groups by approximating sample relationships with a low-rank product used as an affinity. In a synthetic test it gives almost perfect clustering across 10 to 30 samples per subspace, while the full representation reaches around 80% at 30. On Hopkins155, FRR2 has 2.6% mean error versus LRR’s 4.3%; theory assumes clean data and stated conditions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.