Loss Switching Fusion with Similarity Search for Video Classification

Research output: Chapter in Book/Conference paperConference paper

Abstract

From video streaming to security and surveillance applications , video data play an important role in our daily living today. However, managing a large amount of video data and retrieving the most useful information for the user remain a challenging task. In this paper, we propose a novel video classification system that would benefit the scene understanding task. We define our classification problem as classifying background and foreground motions using the same feature representation for outdoor scenes. This means that the feature representation needs to be robust enough and adaptable to different classification tasks. We propose a lightweight Loss Switching Fusion Network (LSFNet) for the fusion of spatiotemporal descriptors and a similarity search scheme with soft voting to boost the classification performance. The proposed system has a variety of potential applications such as content-based video clustering, video filtering, etc. Evaluation results on two private industry datasets show that our system is robust in both classifying different background motions and detecting human motions from these background motions.
Original languageEnglish
Title of host publication2019 IEEE International Conference on Image Processing (ICIP)
Place of PublicationTaiwan
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages974-978
Number of pages5
ISBN (Electronic)978-1-5386-6249-6
DOIs
Publication statusPublished - Sep 2019
EventIEEE International Conference on Image Processing 2019 - Taipai, Taiwan, Province of China
Duration: 22 Sep 201925 Sep 2019
http://2019.ieeeicip.org/

Conference

ConferenceIEEE International Conference on Image Processing 2019
Abbreviated titleICIP2019
CountryTaiwan, Province of China
CityTaipai
Period22/09/1925/09/19
OtherThe 26th IEEE International Conference on Image Processing (ICIP) will be held in Taipei International Convention Center, Taipei, Taiwan, on September 22-25. ICIP is the world’s largest and most comprehensive technical conference focused on image and video processing and computer vision, the conference will feature world-class speakers, tutorials, exhibits and innovative programs that include grand challenges, prominent industry talks, showcases, etc.
Internet address

Fingerprint

Fusion reactions
Video streaming
Industry

Cite this

Wang, L., Huynh, D. Q., & Mansour, M. R. (2019). Loss Switching Fusion with Similarity Search for Video Classification. In 2019 IEEE International Conference on Image Processing (ICIP) (pp. 974-978). Taiwan: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICIP.2019.8803051
Wang, Lei ; Huynh, Du Q. ; Mansour, Moussa Reda. / Loss Switching Fusion with Similarity Search for Video Classification. 2019 IEEE International Conference on Image Processing (ICIP). Taiwan : IEEE, Institute of Electrical and Electronics Engineers, 2019. pp. 974-978
@inproceedings{a1bd4e2e8a64456ebe85d2ed402b6996,
title = "Loss Switching Fusion with Similarity Search for Video Classification",
abstract = "From video streaming to security and surveillance applications , video data play an important role in our daily living today. However, managing a large amount of video data and retrieving the most useful information for the user remain a challenging task. In this paper, we propose a novel video classification system that would benefit the scene understanding task. We define our classification problem as classifying background and foreground motions using the same feature representation for outdoor scenes. This means that the feature representation needs to be robust enough and adaptable to different classification tasks. We propose a lightweight Loss Switching Fusion Network (LSFNet) for the fusion of spatiotemporal descriptors and a similarity search scheme with soft voting to boost the classification performance. The proposed system has a variety of potential applications such as content-based video clustering, video filtering, etc. Evaluation results on two private industry datasets show that our system is robust in both classifying different background motions and detecting human motions from these background motions.",
author = "Lei Wang and Huynh, {Du Q.} and Mansour, {Moussa Reda}",
year = "2019",
month = "9",
doi = "10.1109/ICIP.2019.8803051",
language = "English",
pages = "974--978",
booktitle = "2019 IEEE International Conference on Image Processing (ICIP)",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Wang, L, Huynh, DQ & Mansour, MR 2019, Loss Switching Fusion with Similarity Search for Video Classification. in 2019 IEEE International Conference on Image Processing (ICIP). IEEE, Institute of Electrical and Electronics Engineers, Taiwan, pp. 974-978, IEEE International Conference on Image Processing 2019, Taipai, Taiwan, Province of China, 22/09/19. https://doi.org/10.1109/ICIP.2019.8803051

Loss Switching Fusion with Similarity Search for Video Classification. / Wang, Lei; Huynh, Du Q.; Mansour, Moussa Reda.

2019 IEEE International Conference on Image Processing (ICIP). Taiwan : IEEE, Institute of Electrical and Electronics Engineers, 2019. p. 974-978.

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Loss Switching Fusion with Similarity Search for Video Classification

AU - Wang, Lei

AU - Huynh, Du Q.

AU - Mansour, Moussa Reda

PY - 2019/9

Y1 - 2019/9

N2 - From video streaming to security and surveillance applications , video data play an important role in our daily living today. However, managing a large amount of video data and retrieving the most useful information for the user remain a challenging task. In this paper, we propose a novel video classification system that would benefit the scene understanding task. We define our classification problem as classifying background and foreground motions using the same feature representation for outdoor scenes. This means that the feature representation needs to be robust enough and adaptable to different classification tasks. We propose a lightweight Loss Switching Fusion Network (LSFNet) for the fusion of spatiotemporal descriptors and a similarity search scheme with soft voting to boost the classification performance. The proposed system has a variety of potential applications such as content-based video clustering, video filtering, etc. Evaluation results on two private industry datasets show that our system is robust in both classifying different background motions and detecting human motions from these background motions.

AB - From video streaming to security and surveillance applications , video data play an important role in our daily living today. However, managing a large amount of video data and retrieving the most useful information for the user remain a challenging task. In this paper, we propose a novel video classification system that would benefit the scene understanding task. We define our classification problem as classifying background and foreground motions using the same feature representation for outdoor scenes. This means that the feature representation needs to be robust enough and adaptable to different classification tasks. We propose a lightweight Loss Switching Fusion Network (LSFNet) for the fusion of spatiotemporal descriptors and a similarity search scheme with soft voting to boost the classification performance. The proposed system has a variety of potential applications such as content-based video clustering, video filtering, etc. Evaluation results on two private industry datasets show that our system is robust in both classifying different background motions and detecting human motions from these background motions.

U2 - 10.1109/ICIP.2019.8803051

DO - 10.1109/ICIP.2019.8803051

M3 - Conference paper

SP - 974

EP - 978

BT - 2019 IEEE International Conference on Image Processing (ICIP)

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Taiwan

ER -

Wang L, Huynh DQ, Mansour MR. Loss Switching Fusion with Similarity Search for Video Classification. In 2019 IEEE International Conference on Image Processing (ICIP). Taiwan: IEEE, Institute of Electrical and Electronics Engineers. 2019. p. 974-978 https://doi.org/10.1109/ICIP.2019.8803051