Machine learning

Subnetwork

Detector

Engineering

Artificial intelligence

Transport engineering

Telecommunications

Bounding overwatch

Image (mathematics)

Pedestrian

Computer science

Minimum bounding box

Pedestrian detection

Computer security

Pedestrian detection in a crowd is a very challenging issue. This paper addresses this problem by a novel Non-Maximum Suppression (NMS) algorithm to better refine the bounding boxes given by detectors. The contributions are threefold: (1) we propose adaptive-NMS, which applies a dynamic suppression threshold to an instance, according to the target density; (2) we design an efficient subnetwork to learn density scores, which can be conveniently embedded into both the single-stage and two-stage detectors; and (3) we achieve state of the art results on the CityPersons and CrowdHuman benchmarks.

Adaptive NMS: Refining Pedestrian Detection in a Crowd