Zero-Shot Multi-Animal Tracking in the Wild
Researchers have developed a new zero-shot multi-animal tracking system that leverages vision foundation models, specifically adapting SAM2MOT with Grounding DINO and the Segment Anything Model 2. This method achieves state-of-the-art results across multiple datasets, including Chimp-Act and Bird Flock Tracking, by demonstrating robust generalization across diverse species and environments without requiring retraining. The study also explored the limitations of the newer SAM3 model for this specific application. AI
IMPACT Enables more efficient and generalized animal behavior analysis without dataset-specific tuning.