Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.10530 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912292249010176 |
|---|---|
| author | Nguyen, Quoc-Tien Nguyen, Hong-Hai Huynh, Van-Thong |
| author_facet | Nguyen, Quoc-Tien Nguyen, Hong-Hai Huynh, Van-Thong |
| contents | In this study, we present an approach for efficient spatiotemporal feature extraction using MobileNetV4 and a multi-scale 3D MLP-Mixer-based temporal aggregation module. MobileNetV4, with its Universal Inverted Bottleneck (UIB) blocks, serves as the backbone for extracting hierarchical feature representations from input image sequences, ensuring both computational efficiency and rich semantic encoding. To capture temporal dependencies, we introduce a three-level MLP-Mixer module, which processes spatial features at multiple resolutions while maintaining structural integrity. Experimental results on the ABAW 8th competition demonstrate the effectiveness of our approach, showing promising performance in affective behavior analysis. By integrating an efficient vision backbone with a structured temporal modeling mechanism, the proposed framework achieves a balance between computational efficiency and predictive accuracy, making it well-suited for real-time applications in mobile and embedded computing environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_10530 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Lightweight Models for Emotional Analysis in Video Nguyen, Quoc-Tien Nguyen, Hong-Hai Huynh, Van-Thong Computer Vision and Pattern Recognition Artificial Intelligence In this study, we present an approach for efficient spatiotemporal feature extraction using MobileNetV4 and a multi-scale 3D MLP-Mixer-based temporal aggregation module. MobileNetV4, with its Universal Inverted Bottleneck (UIB) blocks, serves as the backbone for extracting hierarchical feature representations from input image sequences, ensuring both computational efficiency and rich semantic encoding. To capture temporal dependencies, we introduce a three-level MLP-Mixer module, which processes spatial features at multiple resolutions while maintaining structural integrity. Experimental results on the ABAW 8th competition demonstrate the effectiveness of our approach, showing promising performance in affective behavior analysis. By integrating an efficient vision backbone with a structured temporal modeling mechanism, the proposed framework achieves a balance between computational efficiency and predictive accuracy, making it well-suited for real-time applications in mobile and embedded computing environments. |
| title | Lightweight Models for Emotional Analysis in Video |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.10530 |