Self-Supervised Learning on Camera Trap Footage Yields a Strong Universal Face Embedder.

Vladimir Iashin, Horace Lee, Dan Schofield, Andrew Zisserman

Browse the full ICIP paper archive.