Back to Search Start Over

Chained-Tracker: Chaining Paired Attentive Regression Results for End-to-End Joint Multiple-Object Detection and Tracking

Authors :
Fangbin Wan
Changan Wang
Ying Tai
Chengjie Wang
Jilin Li
Jinlong Peng
Feiyue Huang
Yabiao Wang
Yang Wu
Yanwei Fu
Source :
Computer Vision – ECCV 2020 ISBN: 9783030585471, ECCV (4)
Publication Year :
2020
Publisher :
Springer International Publishing, 2020.

Abstract

Existing Multiple-Object Tracking (MOT) methods either follow the tracking-by-detection paradigm to conduct object detection, feature extraction and data association separately, or have two of the three subtasks integrated to form a partially end-to-end solution. Going beyond these sub-optimal frameworks, we propose a simple online model named Chained-Tracker (CTracker), which naturally integrates all the three subtasks into an end-to-end solution (the first as far as we know). It chains paired bounding boxes regression results estimated from overlapping nodes, of which each node covers two adjacent frames. The paired regression is made attentive by object-attention (brought by a detection module) and identity-attention (ensured by an ID verification module). The two major novelties: chained structure and paired attentive regression, make CTracker simple, fast and effective, setting new MOTA records on MOT16 and MOT17 challenge datasets (67.6 and 66.6, respectively), without relying on any extra training data. The source code of CTracker can be found at: github.com/pjl1995/CTracker.

Details

ISBN :
978-3-030-58547-1
ISBNs :
9783030585471
Database :
OpenAIRE
Journal :
Computer Vision – ECCV 2020 ISBN: 9783030585471, ECCV (4)
Accession number :
edsair.doi...........2bdeae1ea854c192a16d3228310b88eb