CrowdHuman: A Benchmark for Detecting Human in a Crowd

30 Apr 2018  ·  Shuai Shao, Zijian Zhao, Boxun Li, Tete Xiao, Gang Yu, Xiangyu Zhang, Jian Sun ·

Human detection has witnessed impressive progress in recent years. However, the occlusion issue of detecting human in highly crowded environments is far from solved. To make matters worse, crowd scenarios are still under-represented in current human detection benchmarks. In this paper, we introduce a new dataset, called CrowdHuman, to better evaluate detectors in crowd scenarios. The CrowdHuman dataset is large, rich-annotated and contains high diversity. There are a total of $470K$ human instances from the train and validation subsets, and $~22.6$ persons per image, with various kinds of occlusions in the dataset. Each human instance is annotated with a head bounding-box, human visible-region bounding-box and human full-body bounding-box. Baseline performance of state-of-the-art detection frameworks on CrowdHuman is presented. The cross-dataset generalization results of CrowdHuman dataset demonstrate state-of-the-art performance on previous dataset including Caltech-USA, CityPersons, and Brainwash without bells and whistles. We hope our dataset will serve as a solid baseline and help promote future research in human detection tasks.

PDF Abstract


Introduced in the Paper:


Used in the Paper:

COCO KITTI CityPersons

Results from the Paper

Ranked #7 on Pedestrian Detection on Caltech (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Pedestrian Detection Caltech FRCNN+FPN-Res50+refined feature map+Crowdhuman Reasonable Miss Rate 3.46 # 7
Pedestrian Detection CityPersons FRCNN+FPN-Res50+refined feature map+Crowdhuman Reasonable MR^-2 10.67 # 11
Object Detection CrowdHuman (full body) Faster RCNN (ResNet50) AP 84.95 # 14
mMR 50.49 # 13


No methods listed for this paper. Add relevant methods here