CA3D: Convolutional-Attentional 3D Nets for Efficient Video Activity Recognition on the Edge
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909623391354880 |
|---|---|
| author | Lagani, Gabriele Falchi, Fabrizio Gennaro, Claudio Amato, Giuseppe |
| author_facet | Lagani, Gabriele Falchi, Fabrizio Gennaro, Claudio Amato, Giuseppe |
| contents | In this paper, we introduce a deep learning solution for video activity recognition that leverages an innovative combination of convolutional layers with a linear-complexity attention mechanism. Moreover, we introduce a novel quantization mechanism to further improve the efficiency of our model during both training and inference. Our model maintains a reduced computational cost, while preserving robust learning and generalization capabilities. Our approach addresses the issues related to the high computing requirements of current models, with the goal of achieving competitive accuracy on consumer and edge devices, enabling smart home and smart healthcare applications where efficiency and privacy issues are of concern. We experimentally validate our model on different established and publicly available video activity recognition benchmarks, improving accuracy over alternative models at a competitive computing cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_19928 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CA3D: Convolutional-Attentional 3D Nets for Efficient Video Activity Recognition on the Edge Lagani, Gabriele Falchi, Fabrizio Gennaro, Claudio Amato, Giuseppe Computer Vision and Pattern Recognition In this paper, we introduce a deep learning solution for video activity recognition that leverages an innovative combination of convolutional layers with a linear-complexity attention mechanism. Moreover, we introduce a novel quantization mechanism to further improve the efficiency of our model during both training and inference. Our model maintains a reduced computational cost, while preserving robust learning and generalization capabilities. Our approach addresses the issues related to the high computing requirements of current models, with the goal of achieving competitive accuracy on consumer and edge devices, enabling smart home and smart healthcare applications where efficiency and privacy issues are of concern. We experimentally validate our model on different established and publicly available video activity recognition benchmarks, improving accuracy over alternative models at a competitive computing cost. |
| title | CA3D: Convolutional-Attentional 3D Nets for Efficient Video Activity Recognition on the Edge |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.19928 |