CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION

Jiawei Ma; Xiaoyu Tao; Jianxing Ma; Xiaopeng Hong; Yihong Gong

doi:10.1109/ICIP42928.2021.9506788

CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION

Jiawei Ma
, Xiaoyu Tao
, Jianxing Ma
, Xiaopeng Hong
, Yihong Gong

College of Artificial Intelligence and Robotics

Xi'an Jiaotong University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

9 Scopus citations

Abstract

Class Incremental Learning (CIL) is a hot topic in machine learning for CNN models to learn new classes incrementally. However, most of the CIL studies are for image classification and object recognition tasks and few CIL studies are available for video action classification. To mitigate this problem, in this paper, we present a new Grow When Required network (GWR) based video CIL framework for action classification. GWR learns knowledge incrementally by modeling the manifold of video frames for each encountered action class in feature space. We also introduce a Knowledge Consolidation (KC) method to separate the feature manifolds of old class and new class and introduce an associative matrix for label prediction. Experimental results on KTH and Weizmann demonstrate the effectiveness of the framework.

Original language	English
Title of host publication	2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings
Publisher	IEEE Computer Society
Pages	504-508
Number of pages	5
ISBN (Electronic)	9781665441155
DOIs	https://doi.org/10.1109/ICIP42928.2021.9506788
State	Published - 2021
Event	28th IEEE International Conference on Image Processing, ICIP 2021 - Anchorage, United States Duration: 19 Sep 2021 → 22 Sep 2021

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
Volume	2021-September
ISSN (Print)	1522-4880

Conference

Conference	28th IEEE International Conference on Image Processing, ICIP 2021
Country/Territory	United States
City	Anchorage
Period	19/09/21 → 22/09/21

Keywords

Class Incremental Learning
Grow When Required network
Video action classification

Access to Document

10.1109/ICIP42928.2021.9506788

Cite this

Ma, J., Tao, X., Ma, J., Hong, X., & Gong, Y. (2021). CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION. In 2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings (pp. 504-508). (Proceedings - International Conference on Image Processing, ICIP; Vol. 2021-September). IEEE Computer Society. https://doi.org/10.1109/ICIP42928.2021.9506788

@inproceedings{fe6f61ecd61045da8a057edc5dba46d4,

title = "CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION",

abstract = "Class Incremental Learning (CIL) is a hot topic in machine learning for CNN models to learn new classes incrementally. However, most of the CIL studies are for image classification and object recognition tasks and few CIL studies are available for video action classification. To mitigate this problem, in this paper, we present a new Grow When Required network (GWR) based video CIL framework for action classification. GWR learns knowledge incrementally by modeling the manifold of video frames for each encountered action class in feature space. We also introduce a Knowledge Consolidation (KC) method to separate the feature manifolds of old class and new class and introduce an associative matrix for label prediction. Experimental results on KTH and Weizmann demonstrate the effectiveness of the framework.",

keywords = "Class Incremental Learning, Grow When Required network, Video action classification",

author = "Jiawei Ma and Xiaoyu Tao and Jianxing Ma and Xiaopeng Hong and Yihong Gong",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 28th IEEE International Conference on Image Processing, ICIP 2021 ; Conference date: 19-09-2021 Through 22-09-2021",

year = "2021",

doi = "10.1109/ICIP42928.2021.9506788",

language = "英语",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "504--508",

booktitle = "2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings",

}

Ma, J, Tao, X, Ma, J, Hong, X & Gong, Y 2021, CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION. in 2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings. Proceedings - International Conference on Image Processing, ICIP, vol. 2021-September, IEEE Computer Society, pp. 504-508, 28th IEEE International Conference on Image Processing, ICIP 2021, Anchorage, United States, 19/09/21. https://doi.org/10.1109/ICIP42928.2021.9506788

CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION. / Ma, Jiawei; Tao, Xiaoyu; Ma, Jianxing et al.
2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings. IEEE Computer Society, 2021. p. 504-508 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2021-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION

AU - Ma, Jiawei

AU - Tao, Xiaoyu

AU - Ma, Jianxing

AU - Hong, Xiaopeng

AU - Gong, Yihong

PY - 2021

Y1 - 2021

N2 - Class Incremental Learning (CIL) is a hot topic in machine learning for CNN models to learn new classes incrementally. However, most of the CIL studies are for image classification and object recognition tasks and few CIL studies are available for video action classification. To mitigate this problem, in this paper, we present a new Grow When Required network (GWR) based video CIL framework for action classification. GWR learns knowledge incrementally by modeling the manifold of video frames for each encountered action class in feature space. We also introduce a Knowledge Consolidation (KC) method to separate the feature manifolds of old class and new class and introduce an associative matrix for label prediction. Experimental results on KTH and Weizmann demonstrate the effectiveness of the framework.

AB - Class Incremental Learning (CIL) is a hot topic in machine learning for CNN models to learn new classes incrementally. However, most of the CIL studies are for image classification and object recognition tasks and few CIL studies are available for video action classification. To mitigate this problem, in this paper, we present a new Grow When Required network (GWR) based video CIL framework for action classification. GWR learns knowledge incrementally by modeling the manifold of video frames for each encountered action class in feature space. We also introduce a Knowledge Consolidation (KC) method to separate the feature manifolds of old class and new class and introduce an associative matrix for label prediction. Experimental results on KTH and Weizmann demonstrate the effectiveness of the framework.

KW - Class Incremental Learning

KW - Grow When Required network

KW - Video action classification

UR - https://www.scopus.com/pages/publications/85123054329

U2 - 10.1109/ICIP42928.2021.9506788

DO - 10.1109/ICIP42928.2021.9506788

M3 - 会议稿件

AN - SCOPUS:85123054329

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 504

EP - 508

BT - 2021 IEEE International Conference on Image Processing, ICIP 2021 - Proceedings

PB - IEEE Computer Society

T2 - 28th IEEE International Conference on Image Processing, ICIP 2021

Y2 - 19 September 2021 through 22 September 2021

ER -

CLASS INCREMENTAL LEARNING FOR VIDEO ACTION CLASSIFICATION

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this