Scale Match for Tiny Person Detection

TL;DR

This paper introduces the TinyPerson benchmark for tiny object detection, identifies scale mismatch as a key challenge, and proposes a Scale Match method to improve detection performance of tiny persons in complex backgrounds.

Contribution

The paper presents a new benchmark for tiny object detection and a simple scale matching technique to enhance feature representation and detection accuracy.

Findings

Significant performance improvements over state-of-the-art detectors

The scale mismatch issue impacts tiny object detection

The TinyPerson benchmark reflects real-world detection challenges

Abstract

Visual object detection has achieved unprecedented ad-vance with the rise of deep convolutional neural networks.However, detecting tiny objects (for example tiny per-sons less than 20 pixels) in large-scale images remainsnot well investigated. The extremely small objects raisea grand challenge about feature representation while themassive and complex backgrounds aggregate the risk offalse alarms. In this paper, we introduce a new benchmark,referred to as TinyPerson, opening up a promising directionfor tiny object detection in a long distance and with mas-sive backgrounds. We experimentally find that the scale mis-match between the dataset for network pre-training and thedataset for detector learning could deteriorate the featurerepresentation and the detectors. Accordingly, we proposea simple yet effective Scale Match approach to align theobject scales between the two datasets for favorable tiny-object representation. Experiments show the significantperformance gain of our proposed approach over state-of-the-art detectors, and the challenging aspects of TinyPersonrelated to real-world scenarios. The TinyPerson benchmarkand the code for our approach will be publicly available(https://github.com/ucas-vg/TinyBenchmark).(Attention: evaluation rules of AP have updated in benchmark after this paper accepted, So this paper use old rules. we will keep old rules of AP in benchmark, but we recommand the new and we will use the new in latter research.)