Ten Years of Pedestrian Detection, What Have We Learned?

Abstract

Paper-by-paper results make it easy to miss the forest for the trees.We analyse the remarkable progress of the last decade by discussing the main ideas explored in the 40+ detectors currently present in the Caltech pedestrian detection benchmark. We observe that there exist three families of approaches, all currently reaching similar detection quality. Based on our analysis, we study the complementarity of the most promising ideas by combining multiple published strategies. This new decision forest detector achieves the current best known performance on the challenging Caltech-USA dataset.

To appear in ECCV 2014 CVRSUAD workshop proceedings

Cited by in corpus (13)

• Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking
• Perceptual Generative Adversarial Networks for Small Object Detection
• CityPersons: A Diverse Dataset for Pedestrian Detection
• Filtered Channel Features for Pedestrian Detection
• Multispectral Deep Neural Networks for Pedestrian Detection
• Multi-camera Multi-Object Tracking
• ViTAA: Visual-Textual Attributes Alignment in Person Search by Natural Language
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Crowd Counting Using Scale-Aware Attention Networks
• Is Faster R-CNN Doing Well for Pedestrian Detection?
• Vulnerable road user detection: state-of-the-art and open challenges
• Pedestrian Detection aided by Deep Learning Semantic Tasks
• Addressing Ambiguity in Multi-target Tracking by Hierarchical Strategy