Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Long & Short Memory Balancing In Visual Co-tracking using Q-learning

Abstract: 

Employing one or more additional classifiers to break the self-learning loop in tracing-by-detection has gained considerable attention. Most of such trackers merely utilize the redundancy to address the accumulating label error in the tracking loop, and suffer from high computational complexity as well as tracking challenges that may interrupt all classifiers (e.g. temporal occlusions). We propose the active co-tracking framework, in which the main classifier of the tracker labels samples of the video sequence, and only consults auxiliary classifier when it is uncertain. Based on the source of the uncertainty and the differences of two classifiers (e.g. accuracy, speed, update frequency, etc.), different policies should be taken to exchange information between two classifiers. Here, we introduce a reinforcement learning approach to find the appropriate policy by considering the state of tracker in a specific sequence. The proposed method yields promising results in comparison to the best tracking-by-detection approaches.

up
0 users have voted:

Paper Details

Authors:
Maryam Sadat Mirzaei, Shigeyuki Oba
Submitted On:
20 September 2019 - 9:57am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Kourosh Meshgi, Maryam Sadat Mirzaei
Paper Code:
WP.PC.2
Document Year:
2019
Cite

Document Files

meshgi-icip poster v2.pdf

(11)

Subscribe

[1] Maryam Sadat Mirzaei, Shigeyuki Oba, "Long & Short Memory Balancing In Visual Co-tracking using Q-learning", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4780. Accessed: Oct. 18, 2019.
@article{4780-19,
url = {http://sigport.org/4780},
author = {Maryam Sadat Mirzaei; Shigeyuki Oba },
publisher = {IEEE SigPort},
title = {Long & Short Memory Balancing In Visual Co-tracking using Q-learning},
year = {2019} }
TY - EJOUR
T1 - Long & Short Memory Balancing In Visual Co-tracking using Q-learning
AU - Maryam Sadat Mirzaei; Shigeyuki Oba
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4780
ER -
Maryam Sadat Mirzaei, Shigeyuki Oba. (2019). Long & Short Memory Balancing In Visual Co-tracking using Q-learning. IEEE SigPort. http://sigport.org/4780
Maryam Sadat Mirzaei, Shigeyuki Oba, 2019. Long & Short Memory Balancing In Visual Co-tracking using Q-learning. Available at: http://sigport.org/4780.
Maryam Sadat Mirzaei, Shigeyuki Oba. (2019). "Long & Short Memory Balancing In Visual Co-tracking using Q-learning." Web.
1. Maryam Sadat Mirzaei, Shigeyuki Oba. Long & Short Memory Balancing In Visual Co-tracking using Q-learning [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4780