PriVi: Towards A General-Purpose Video Model For Primate Behavior In The Wild
This addresses the problem of limited generalization in computer vision for primate behavior analysis, benefiting researchers in cognition, evolution, and conservation, though it is incremental as it adapts existing methods to a new domain.
The paper tackled the limitation of human-centric pretrained models for analyzing primate behavior by introducing PriVi, a large-scale primate-centric video pretraining dataset, and showed that pretraining V-JEPA on it outperforms prior work across four benchmark datasets, improving data efficiency and generalization.
Non-human primates are our closest living relatives, and analyzing their behavior is central to research in cognition, evolution, and conservation. Computer vision could greatly aid this research, but existing methods often rely on human-centric pretrained models and focus on single datasets, which limits generalization. We address this limitation by shifting from a model-centric to a data-centric approach and introduce PriVi, a large-scale primate-centric video pretraining dataset. PriVi contains 424 hours of curated video, combining 174 hours from behavioral research across 11 settings with 250 hours of diverse web-sourced footage, assembled through a scalable data curation pipeline. We pretrain V-JEPA, a large-scale video model, on PriVi to learn primate-specific representations and evaluate it using a lightweight frozen classifier. Across four benchmark datasets, ChimpACT, BaboonLand, PanAf500, and ChimpBehave, our approach consistently outperforms prior work, including fully finetuned baselines, and scales favorably with fewer labels. These results demonstrate that primate-centric pretraining substantially improves data efficiency and generalization, making it a promising approach for low-label applications. Code, models, and the majority of the dataset will be made available.