WEKO3
アイテム
Video Classification Using Smooth Approximation of Hard-assignment Encoding
https://ipsj.ixsq.nii.ac.jp/records/238008
https://ipsj.ixsq.nii.ac.jp/records/238008d4f1db92-d7a9-47cb-ae08-6aaee0f87ba3
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
IPSJ-JNL6508011.pdf (758.7 kB)
2026年8月15日からダウンロード可能です。
|
Copyright (c) 2024 by the Information Processing Society of Japan
|
|
| 非会員：¥0, IPSJ:学会員：¥0, 論文誌:会員：¥0, DLIB:会員：¥0 | ||
| Item type | Journal(1) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2024-08-15 | |||||||||
| タイトル | ||||||||||
| タイトル | Video Classification Using Smooth Approximation of Hard-assignment Encoding | |||||||||
| タイトル | ||||||||||
| 言語 | en | |||||||||
| タイトル | Video Classification Using Smooth Approximation of Hard-assignment Encoding | |||||||||
| 言語 | ||||||||||
| 言語 | eng | |||||||||
| キーワード | ||||||||||
| 主題Scheme | Other | |||||||||
| 主題 | [一般論文] convolutional neural network (CNN), columbia consumer video (CCV) dataset, Kinetics-400 dataset, hard assignment, vector of locally aggregated descriptor (VLAD), smooth approximation | |||||||||
| 資源タイプ | ||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||||||
| 資源タイプ | journal article | |||||||||
| 著者所属 | ||||||||||
| Department of Computer Science, Faculty of Mathematical Sciences and Computer, Kharazmi University | ||||||||||
| 著者所属 | ||||||||||
| Department of Computer Science, Faculty of Mathematical Sciences and Computer, Kharazmi University | ||||||||||
| 著者所属(英) | ||||||||||
| en | ||||||||||
| Department of Computer Science, Faculty of Mathematical Sciences and Computer, Kharazmi University | ||||||||||
| 著者所属(英) | ||||||||||
| en | ||||||||||
| Department of Computer Science, Faculty of Mathematical Sciences and Computer, Kharazmi University | ||||||||||
| 著者名 |
Mohammad, Soltanian
× Mohammad, Soltanian
× Keivan, Borna
|
|||||||||
| 著者名(英) |
Mohammad, Soltanian
× Mohammad, Soltanian
× Keivan, Borna
|
|||||||||
| 論文抄録 | ||||||||||
| 内容記述タイプ | Other | |||||||||
| 内容記述 | The classification of encoded frame level descriptors is a common approach to video event recognition. However, the structural incorporation of visual temporal clues to the encoding process is often ignored, resulting in reduced recognition accuracy. In this paper, a spatio-temporal video encoding method is proposed that improves the accuracy of video event recognition. By fine-tuning the Convolutional Neural Network (CNN) concept score extractors, pre-trained on ImageNet, frame level descriptors are computed. The descriptors are then encoded and normalized and are fed to the classifier to discriminate between the video events. The main contribution here is to use the temporal dimension of video signals to construct a spatio-temporal vector of locally aggregated descriptors (VLAD) encoding scheme. The proposed encoding is shown to be in the form of a non-convex constrained optimization problem with l0 norm terms, which is transformed, by a Gaussian approximation, into a smoothed version. This makes the cost function differentiable and overcomes the non-smoothness challenge. Compared to the state-of-the-art video event recognition schemes, the proposed method achieves comparable performance as examined over two public datasets: Columbia consumer video (CCV) and Kinetics-400. ------------------------------ This is a preprint of an article intended for publication Journal of Information Processing(JIP). This preprint should not be cited. This article should be cited as: Journal of Information Processing Vol.32(2024) (online) DOI http://dx.doi.org/10.2197/ipsjjip.32.641 ------------------------------ |
|||||||||
| 論文抄録(英) | ||||||||||
| 内容記述タイプ | Other | |||||||||
| 内容記述 | The classification of encoded frame level descriptors is a common approach to video event recognition. However, the structural incorporation of visual temporal clues to the encoding process is often ignored, resulting in reduced recognition accuracy. In this paper, a spatio-temporal video encoding method is proposed that improves the accuracy of video event recognition. By fine-tuning the Convolutional Neural Network (CNN) concept score extractors, pre-trained on ImageNet, frame level descriptors are computed. The descriptors are then encoded and normalized and are fed to the classifier to discriminate between the video events. The main contribution here is to use the temporal dimension of video signals to construct a spatio-temporal vector of locally aggregated descriptors (VLAD) encoding scheme. The proposed encoding is shown to be in the form of a non-convex constrained optimization problem with l0 norm terms, which is transformed, by a Gaussian approximation, into a smoothed version. This makes the cost function differentiable and overcomes the non-smoothness challenge. Compared to the state-of-the-art video event recognition schemes, the proposed method achieves comparable performance as examined over two public datasets: Columbia consumer video (CCV) and Kinetics-400. ------------------------------ This is a preprint of an article intended for publication Journal of Information Processing(JIP). This preprint should not be cited. This article should be cited as: Journal of Information Processing Vol.32(2024) (online) DOI http://dx.doi.org/10.2197/ipsjjip.32.641 ------------------------------ |
|||||||||
| 書誌レコードID | ||||||||||
| 収録物識別子タイプ | NCID | |||||||||
| 収録物識別子 | AN00116647 | |||||||||
| 書誌情報 |
情報処理学会論文誌 巻 65, 号 8, 発行日 2024-08-15 |
|||||||||
| ISSN | ||||||||||
| 収録物識別子タイプ | ISSN | |||||||||
| 収録物識別子 | 1882-7764 | |||||||||
| 公開者 | ||||||||||
| 言語 | ja | |||||||||
| 出版者 | 情報処理学会 | |||||||||
