Abstract
This work focuses on per-video unsupervised action segmentation, which is of interest to applications where storing large datasets is either not possible, or nor permitted. We propose to segment videos by learning in deep kernel space, to approximate the underlying frame distribution, as closely as possible. To define this closeness metric between the original video distribution and its approximation, we rely on maximum mean discrepancy (MMD) which is a geometry-preserving metric in distribution space, and thus gives more reliable estimates. Moreover, unlike the commonly used optimal transport metric, MMD is both easier to optimize, and faster. We choose to use neural tangent kernels (NTKs) to define the kernel space where MMD operates, because of their improved descriptive power as opposed to fixed kernels. And, also, because NTKs sidestep the trivial solution, when jointly learning the inputs (video approximation) and the kernel function. Finally, we show competitive results when compared to state-of-the-art per-video methods, on six standard benchmarks. Additionally, our method has higher scores than prior agglomerative work, when the number of segments is unknown.
| Original language | English |
|---|---|
| Title of host publication | Pattern recognition |
| Subtitle of host publication | 28th International conference, ICPR 2026, Lyon, France, August 17–22, 2026, Proceedings, Part I |
| Editors | Maria De Marsico, Tin Kam Ho, Frederic Jurie, Cheng-Lin Liu, Daniel Lopresti, Ingela Nyström, Jean-Marc Ogier, Arun Ross, Liang Wang |
| Publisher | Springer Cham |
| Pages | 32-47 |
| Number of pages | 16 |
| Edition | 1st |
| ISBN (Electronic) | 9783032313355 |
| ISBN (Print) | 9783032313348 |
| DOIs | |
| Publication status | Published - Aug 2026 |
| Event | 28th International conference on pattern recognition - Lyon, France Duration: 17 Aug 2026 → 22 Aug 2026 Conference number: 28th |
Publication series
| Name | Lecture notes in computer science |
|---|---|
| Publisher | Springer Nature |
| Volume | 16812 |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Conference
| Conference | 28th International conference on pattern recognition |
|---|---|
| Abbreviated title | ICPR 2026 |
| Country/Territory | France |
| City | Lyon |
| Period | 17/08/26 → 22/08/26 |
Keywords
- unsupervised action segmentation
- MMD
- infinite NTK
Fingerprint
Dive into the research topics of 'Deep kernel video approximation for unsupervised action segmentation'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver