Deep Occlusion Reasoning for Multi-Camera Multi-Target Detection

2017/04/19 by Baqué, Pierre, Fleuret, François, Fua, Pascal · 5 citations
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1704.05775

Abstract

People detection in single 2D images has improved greatly in recent years. However, comparatively little of this progress has percolated into multi-camera multi-people tracking algorithms, whose performance still degrades severely when scenes become very crowded. In this work, we introduce a new architecture that combines Convolutional Neural Nets and Conditional Random Fields to explicitly model those ambiguities. One of its key ingredients are high-order CRF terms that model potential occlusions and give our approach its robustness even when many people are present. Our model is trained end-to-end and we show that it outperforms several state-of-art algorithms on challenging scenes.

Cited by

• 3D Ground Truth Reconstruction from Multi-Camera Annotations Using UKF
• Active View Selection for Scene-level Multi-view Crowd Counting and Localization with Limited Labels
• Sparse BEV Fusion with Self-View Consistency for Multi-View Detection and Tracking
• MVTrajecter: Multi-View Pedestrian Tracking with Trajectory Motion Cost and Trajectory Appearance Cost
• MSMVD: Exploiting Multi-scale Image Features via Multi-scale BEV Features for Multi-view Pedestrian Detection

Related

• Deep Multi-camera People Detection
• Towards Dense People Detection with Deep Learning and Depth images
• Multi-View People Detection in Large Scenes via Supervised View-Wise Contribution Weighting
• Multi-Person Tracking by Multicut and Deep Matching
• Multi-view Tracking Using Weakly Supervised Human Motion Prediction