HAJJv2-CrowdCount: Zero-Shot Benchmark for Dense Crowd Counting
Provides a public benchmark and reveals that point-based counting is more reliable than detection/segmentation methods in extreme occlusion, which is critical for Hajj crowd management.
The authors introduce HAJJv2-CrowdCount, a per-second annotated benchmark for dense crowd counting in Hajj videos, and evaluate three zero-shot counting methods. SAM3Count achieves the lowest overall MAE (70.4), but on the densest frames the point-based counter APGCC degrades more gracefully (MAE 114.9) than detection/segmentation-based methods (MAE >300).
Automated crowd counting in Hajj video is difficult not because current models lack capacity, but because the footage violates the assumptions those models were built on: cameras observe the crowd from steep, near-vertical angles, individuals occlude one another extensively, and a single frame can contain well over a thousand people. Benchmarks that test crowd counting in such an environment are either private or not detailed per second. We revisit the HAJJv2 dataset and contribute HAJJv2-CrowdCount: per-second human-annotated crowd counts for its testing videos. Using these annotations, we benchmark three recent zero-shot counting paradigms: an open-vocabulary detector (YOLO-World), a point-based counter (APGCC), and a promptable segmentation-based counter (SAM3Count). SAM3Count attains the lowest overall mean absolute error (MAE 70.4, 95% CI 56.0-86.1), ahead of YOLO-World (92.0) and APGCC (152.9). This ordering reverses, however, in the regime most relevant to deployment: on the densest frames, the detection- and segmentation-based counters both degrade sharply (MAE exceeding 300), while the point-based counter degrades far more gracefully (MAE 114.9). This inversion is decision-relevant for Hajj crowd management, where reliable counts are needed most precisely in the densest and most occluded scenes. The annotations are released to support reproduction and extension of these results.