Quantifying and Learning Static vs. Dynamic Information in Deep Spatiotemporal Networks
There is limited understanding of the information captured by deep spatiotemporal models in their intermediate representations. For example, while evidence suggests that action recognition algorithms are heavily influenced by visual appearance in single frames, no quantitative methodology exists for...
Ausführliche Beschreibung
Bibliographische Detailangaben
| Veröffentlicht in: | IEEE transactions on pattern analysis and machine intelligence. - 1979. - 47(2025), 1 vom: 01. Jan., Seite 190-205
|
|---|
| 1. Verfasser: |
Kowal, Matthew
(VerfasserIn) |
|---|
| Weitere Verfasser: |
Siam, Mennatullah,
Islam, Md Amirul,
Bruce, Neil D B,
Wildes, Richard P,
Derpanis, Konstantinos G |
|---|
| Format: | Online-Aufsatz
|
|---|
| Sprache: | English |
|---|
| Veröffentlicht: |
2025
|
|---|
| Zugriff auf das übergeordnete Werk: | IEEE transactions on pattern analysis and machine intelligence
|
|---|
| Schlagworte: | Journal Article |
|---|