A self-supervised vision transformer trained on mined unlabeled chimpanzee faces outperforms supervised baselines for open-set re-identification on wild footage.
We mine two balanced corpora with the Data Engine in §3.2: PanAf-20K and Loma Mountains each contribute ∼314k high-quality face crops
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Self-supervised Learning on Camera Trap Footage Yields a Strong Universal Face Embedder
A self-supervised vision transformer trained on mined unlabeled chimpanzee faces outperforms supervised baselines for open-set re-identification on wild footage.