ARGUSTRACK cuts multi-camera annotation time using BEV plane

Multi-Camera Multi-Target (MCMT) tracking is essential for applications like autonomous driving and animal behavior monitoring, but obtaining annotated multi-view data remains a major bottleneck. Existing annotation tools mostly support single-camera workflows or rely on LiDAR sensors, making cross-view labeling tedious and impractical for camera-only setups. Researchers from arXiv present ARGUSTRACK, a multi-camera annotation system that solves this by allowing annotators to work directly on a bird's-eye-view (BEV) plane. Given calibrated camera parameters, a single ground-plane annotation is automatically projected into 2D bounding boxes across all relevant views, ensuring identity consistency without any manual cross-view alignment.

To further accelerate the labeling process, ARGUSTRACK incorporates two complementary mechanisms. First, a Temporal Aware module propagates annotations from preceding frames to initialize new ones, requiring only minor positional adjustments. Second, a Multi-camera Semi-annotation module leverages off-the-shelf 2D detectors combined with foot-point estimation to automatically generate candidate BEV positions for annotator verification. The system was evaluated through a pilot study on multi-camera broiler tracking, demonstrating that it substantially reduces annotation time compared to conventional single-camera labeling workflows. This tool could significantly speed up dataset creation for multi-view tracking tasks in robotics, surveillance, and ecology.