Reproduce “RFOP: Rethinking Fusion and Orthogonal Projection for Face-Voice Association” 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
Abdul Hannan, Furqan Malik, Hina Jabbar, Syed Suleman Sadiq, Mubashir Noman
RFOP addresses face-voice verification by aligning face and voice features in a shared space and combining them with a fusion module. On FAME’26 dataset-V3, English-trained RFOP reports EERs of 25.4 on English tests and 41.1 on German tests, versus 35.3 and 48.0 for FOP. Overall scores are 33.1 and 41.5; aggregation is unspecified, and these results do not establish transfer beyond this dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.