Deep attentive tracking via reciprocative learning

Shi Pu, Yibing Song, Chao Ma, Honggang Zhang, Ming Hsuan Yang

Research output: Contribution to journalConference article

7 Citations (Scopus)

Abstract

Visual attention, derived from cognitive neuroscience, facilitates human perception on the most pertinent subset of the sensory data. Recently, significant efforts have been made to exploit attention schemes to advance computer vision systems. For visual tracking, it is often challenging to track target objects undergoing large appearance changes. Attention maps facilitate visual tracking by selectively paying attention to temporal robust features. Existing tracking-by-detection approaches mainly use additional attention modules to generate feature weights as the classifiers are not equipped with such mechanisms. In this paper, we propose a reciprocative learning algorithm to exploit visual attention for training deep classifiers. The proposed algorithm consists of feed-forward and backward operations to generate attention maps, which serve as regularization terms coupled with the original classification loss function for training. The deep classifier learns to attend to the regions of target objects robust to appearance changes. Extensive experiments on large-scale benchmark datasets show that the proposed attentive tracking method performs favorably against the state-of-the-art approaches.

Original languageEnglish
Pages (from-to)1931-1941
Number of pages11
JournalAdvances in Neural Information Processing Systems
Volume2018-December
Publication statusPublished - 2018 Jan 1
Event32nd Conference on Neural Information Processing Systems, NeurIPS 2018 - Montreal, Canada
Duration: 2018 Dec 22018 Dec 8

Fingerprint

Classifiers
Learning algorithms
Computer vision
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Pu, Shi ; Song, Yibing ; Ma, Chao ; Zhang, Honggang ; Yang, Ming Hsuan. / Deep attentive tracking via reciprocative learning. In: Advances in Neural Information Processing Systems. 2018 ; Vol. 2018-December. pp. 1931-1941.
@article{e7ba615fbe3245dca6933e6679fb2a0e,
title = "Deep attentive tracking via reciprocative learning",
abstract = "Visual attention, derived from cognitive neuroscience, facilitates human perception on the most pertinent subset of the sensory data. Recently, significant efforts have been made to exploit attention schemes to advance computer vision systems. For visual tracking, it is often challenging to track target objects undergoing large appearance changes. Attention maps facilitate visual tracking by selectively paying attention to temporal robust features. Existing tracking-by-detection approaches mainly use additional attention modules to generate feature weights as the classifiers are not equipped with such mechanisms. In this paper, we propose a reciprocative learning algorithm to exploit visual attention for training deep classifiers. The proposed algorithm consists of feed-forward and backward operations to generate attention maps, which serve as regularization terms coupled with the original classification loss function for training. The deep classifier learns to attend to the regions of target objects robust to appearance changes. Extensive experiments on large-scale benchmark datasets show that the proposed attentive tracking method performs favorably against the state-of-the-art approaches.",
author = "Shi Pu and Yibing Song and Chao Ma and Honggang Zhang and Yang, {Ming Hsuan}",
year = "2018",
month = "1",
day = "1",
language = "English",
volume = "2018-December",
pages = "1931--1941",
journal = "Advances in Neural Information Processing Systems",
issn = "1049-5258",

}

Pu, S, Song, Y, Ma, C, Zhang, H & Yang, MH 2018, 'Deep attentive tracking via reciprocative learning', Advances in Neural Information Processing Systems, vol. 2018-December, pp. 1931-1941.

Deep attentive tracking via reciprocative learning. / Pu, Shi; Song, Yibing; Ma, Chao; Zhang, Honggang; Yang, Ming Hsuan.

In: Advances in Neural Information Processing Systems, Vol. 2018-December, 01.01.2018, p. 1931-1941.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Deep attentive tracking via reciprocative learning

AU - Pu, Shi

AU - Song, Yibing

AU - Ma, Chao

AU - Zhang, Honggang

AU - Yang, Ming Hsuan

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Visual attention, derived from cognitive neuroscience, facilitates human perception on the most pertinent subset of the sensory data. Recently, significant efforts have been made to exploit attention schemes to advance computer vision systems. For visual tracking, it is often challenging to track target objects undergoing large appearance changes. Attention maps facilitate visual tracking by selectively paying attention to temporal robust features. Existing tracking-by-detection approaches mainly use additional attention modules to generate feature weights as the classifiers are not equipped with such mechanisms. In this paper, we propose a reciprocative learning algorithm to exploit visual attention for training deep classifiers. The proposed algorithm consists of feed-forward and backward operations to generate attention maps, which serve as regularization terms coupled with the original classification loss function for training. The deep classifier learns to attend to the regions of target objects robust to appearance changes. Extensive experiments on large-scale benchmark datasets show that the proposed attentive tracking method performs favorably against the state-of-the-art approaches.

AB - Visual attention, derived from cognitive neuroscience, facilitates human perception on the most pertinent subset of the sensory data. Recently, significant efforts have been made to exploit attention schemes to advance computer vision systems. For visual tracking, it is often challenging to track target objects undergoing large appearance changes. Attention maps facilitate visual tracking by selectively paying attention to temporal robust features. Existing tracking-by-detection approaches mainly use additional attention modules to generate feature weights as the classifiers are not equipped with such mechanisms. In this paper, we propose a reciprocative learning algorithm to exploit visual attention for training deep classifiers. The proposed algorithm consists of feed-forward and backward operations to generate attention maps, which serve as regularization terms coupled with the original classification loss function for training. The deep classifier learns to attend to the regions of target objects robust to appearance changes. Extensive experiments on large-scale benchmark datasets show that the proposed attentive tracking method performs favorably against the state-of-the-art approaches.

UR - http://www.scopus.com/inward/record.url?scp=85060857631&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060857631&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85060857631

VL - 2018-December

SP - 1931

EP - 1941

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

SN - 1049-5258

ER -