Occluded Pedestrian Detection with Visible IoU and Box Sign Predictor

Abstract

Training a robust classifier and an accurate box regressor are difficult for occluded pedestrian detection. Traditionally adopted Intersection over Union (IoU) measurement does not consider the occluded region of the object and leads to improper training samples. To address such issue, a modification called visible IoU is proposed in this paper to explicitly incorporate the visible ratio in selecting samples. Then a newly designed box sign predictor is placed in parallel with box regressor to separately predict the moving direction of training samples. It leads to higher localization accuracy by introducing sign prediction loss during training and sign refining in testing. Following these novelties, we obtain state-of-the-art performance on CityPersons benchmark for occluded pedestrian detection.

arXiv commentThe 26th IEEE International Conference on Image Processing (ICIP)

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Small-scale Pedestrian Detection Based on Somatic Topology Localization and Temporal Feature Aggregation
• Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd