DeST: A Decoupled Spatio-Temporal Framework for Action Segmentation.
Saved in:
| Title: | DeST: A Decoupled Spatio-Temporal Framework for Action Segmentation. |
|---|---|
| Authors: | Li, Yunheng1 (AUTHOR) yunhengli@mail.nankai.edu.cn, Li, Zhong-Yu1 (AUTHOR) lizhongyu@mail.nankai.edu.cn, Gao, Shanghua1 (AUTHOR), Wang, Qilong2 (AUTHOR), Hou, Qibin1,3 (AUTHOR) houqb@nankai.edu.cn, Cheng, Ming-Ming1,3 (AUTHOR) cmm@nankai.edu.cn |
| Source: | International Journal of Computer Vision. May2026, Vol. 134 Issue 5, p1-21. 21p. |
| Subjects: | Spatiotemporal processes, Movement sequences, Computer science, Software frameworks |
| Abstract: | Effectively modeling discriminative spatio-temporal information is essential for segmenting activities in long action sequences. However, we observe that existing methods are limited in weak spatio-temporal modeling capability due to two forms of coupled modeling: (i) Cascaded interaction couples spatial and temporal modeling, which over-smooths motion modeling over the long sequence, and (ii) Joint-shared temporal modeling adopts shared weights to model each joint, ignoring the distinct motion patterns of different joints. In this paper, we present a Decoupled Spatio-Temporal Framework (DeST) to address the above issues. Firstly, we decouple the cascaded spatio-temporal interaction to avoid stacking multiple spatio-temporal blocks, while achieving sufficient spatio-temporal interaction. Specifically, DeST performs once unified spatial modeling and divides the spatial features into different groups of sub-features, which then adaptively interact with temporal features from different layers. Since the different sub-features contain distinct spatial semantics, the model could learn better interaction patterns at each layer. Meanwhile, inspired by the fact that different joints move at different speeds, we propose joint-decoupled temporal modeling, which employs independent trainable weights to capture distinctive temporal features of each joint. On four large-scale benchmarks of different scenes, DeST significantly outperforms current state-of-the-art methods with less computational complexity. Our code is available at: https://github.com/lyhisme/DeST. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 192788063 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: DeST: A Decoupled Spatio-Temporal Framework for Action Segmentation. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Li%2C+Yunheng%22">Li, Yunheng</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yunhengli@mail.nankai.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Li%2C+Zhong-Yu%22">Li, Zhong-Yu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> lizhongyu@mail.nankai.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Gao%2C+Shanghua%22">Gao, Shanghua</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Qilong%22">Wang, Qilong</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hou%2C+Qibin%22">Hou, Qibin</searchLink><relatesTo>1,3</relatesTo> (AUTHOR)<i> houqb@nankai.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Cheng%2C+Ming-Ming%22">Cheng, Ming-Ming</searchLink><relatesTo>1,3</relatesTo> (AUTHOR)<i> cmm@nankai.edu.cn</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22International+Journal+of+Computer+Vision%22">International Journal of Computer Vision</searchLink>. May2026, Vol. 134 Issue 5, p1-21. 21p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Spatiotemporal+processes%22">Spatiotemporal processes</searchLink><br /><searchLink fieldCode="DE" term="%22Movement+sequences%22">Movement sequences</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+science%22">Computer science</searchLink><br /><searchLink fieldCode="DE" term="%22Software+frameworks%22">Software frameworks</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Effectively modeling discriminative spatio-temporal information is essential for segmenting activities in long action sequences. However, we observe that existing methods are limited in weak spatio-temporal modeling capability due to two forms of coupled modeling: (i) Cascaded interaction couples spatial and temporal modeling, which over-smooths motion modeling over the long sequence, and (ii) Joint-shared temporal modeling adopts shared weights to model each joint, ignoring the distinct motion patterns of different joints. In this paper, we present a Decoupled Spatio-Temporal Framework (DeST) to address the above issues. Firstly, we decouple the cascaded spatio-temporal interaction to avoid stacking multiple spatio-temporal blocks, while achieving sufficient spatio-temporal interaction. Specifically, DeST performs once unified spatial modeling and divides the spatial features into different groups of sub-features, which then adaptively interact with temporal features from different layers. Since the different sub-features contain distinct spatial semantics, the model could learn better interaction patterns at each layer. Meanwhile, inspired by the fact that different joints move at different speeds, we propose joint-decoupled temporal modeling, which employs independent trainable weights to capture distinctive temporal features of each joint. On four large-scale benchmarks of different scenes, DeST significantly outperforms current state-of-the-art methods with less computational complexity. Our code is available at: https://github.com/lyhisme/DeST. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=192788063 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s11263-026-02797-0 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 21 StartPage: 1 Subjects: – SubjectFull: Spatiotemporal processes Type: general – SubjectFull: Movement sequences Type: general – SubjectFull: Computer science Type: general – SubjectFull: Software frameworks Type: general Titles: – TitleFull: DeST: A Decoupled Spatio-Temporal Framework for Action Segmentation. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Li, Yunheng – PersonEntity: Name: NameFull: Li, Zhong-Yu – PersonEntity: Name: NameFull: Gao, Shanghua – PersonEntity: Name: NameFull: Wang, Qilong – PersonEntity: Name: NameFull: Hou, Qibin – PersonEntity: Name: NameFull: Cheng, Ming-Ming IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 05 Text: May2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 09205691 Numbering: – Type: volume Value: 134 – Type: issue Value: 5 Titles: – TitleFull: International Journal of Computer Vision Type: main |
| ResultId | 1 |
