scholarly journals Multi-Person Tracking Based on Faster R-CNN and Deep Appearance Features

2019 ◽  
Author(s):  
Gulraiz Khan ◽  
Zeeshan Tariq ◽  
Muhammad Usman Ghani Khan
Keyword(s):  
Person Tracking ◽  
Multi Person Tracking

scholarly journals A Graphical Social Topology Model for RGB-D Multi-Person Tracking

2021 ◽  
pp. 1-1
Author(s):  
Shan Gao ◽  
Qixiang Ye ◽  
Li Liu ◽  
Arjan Kuijper ◽  
Xiangyang Ji
Keyword(s):  
Person Tracking ◽  
Topology Model ◽  
Multi Person Tracking ◽  
Social Topology

scholarly journals Multi-person Tracking by Multicut and Deep Matching

2016 ◽  
pp. 100-111 ◽  
Author(s):  
Siyu Tang ◽  
Bjoern Andres ◽  
Mykhaylo Andriluka ◽  
Bernt Schiele
Keyword(s):  
Person Tracking ◽  
Multi Person Tracking

scholarly journals Instance-aware representation learning and association for online multi-person tracking

2019 ◽  
Vol 94 ◽  
pp. 25-34 ◽  
Author(s):  
Hefeng Wu ◽  
Yafei Hu ◽  
Keze Wang ◽  
Hanhui Li ◽  
Lin Nie ◽  
...  
Keyword(s):  
Representation Learning ◽  
Person Tracking ◽  
Multi Person Tracking

scholarly journals Online semi-supervised multi-person tracking with gaussian process regression

2019 ◽  
Vol 277 ◽  
pp. 01003 ◽  
Author(s):  
Baobing Zhang ◽  
Zhengwen Huang ◽  
Babak H. Rahi ◽  
Qicong Wang ◽  
Maozhen Li
Keyword(s):  
Prior Information ◽  
Learning Strategy ◽  
Gaussian Process Regression ◽  
Batch Processing ◽  
Current Frame ◽  
Lighting Condition ◽  
New Approach ◽  
Person Tracking ◽  
The Arts ◽  
Multi Person Tracking

Most existing multi-person tracking approaches are affected by lighting condition, pedestrian pose change abruptly, scale changes, realtime processing to name a few, resulting in detection error, drift and other issues. To cope with this challenge, we propose an enhanced multi-person framework by introducing a new observation model, which adaptively updates fully online to avoid the loss of sample diversity and learning in a semi-supervised manner. We fuse prior information for tracking decision, meanwhile extracted knowledge from current frame is used to assist to make tracking decision, which can be viewed as a transfer learning strategy, and both aspects can ameliorate the tendency to drift. The new approach does not need any calibration or batch processing. Experimental results show that the approach yields comparable or better performance in comparison with the state-of-the-arts, which do calibration or batch processing.

Sign in / Sign up

Export Citation Format

Share Document