Scale Match for Tiny Person Detection

Abstract

Visual object detection has achieved unprecedented advance with the rise of deep convolutional neural networks.However, detecting tiny objects (for example tiny persons less than 20 pixels) in large-scale images remains challenging. The extremely small objects raise a grand challenge about feature representation while the massive and complex backgrounds aggregates the risk of false detections. In this paper, we introduce a new benchmark, referred to as TinyPerson, opening up a promising direction for tiny object detection in a long distance and with massive back-grounds. We experimentally find that the scale mismatch be-tween the dataset for network pretraining and the dataset for detector learning could deteriorate the feature representation and the detectors. Accordingly, we propose a simple yet effective Scale Match approach to align the object scales between the two datasets for favorable tiny-object representation. Experiments show the significant performance gain of our proposed approach over state-of-the-art detectors, and the challenging aspects of TinyPerson related to real-world scenarios. The TinyPerson benchmark and the code for our approach will be publicly available.