Using Cross-Domain Detection Loss to Infer Multi-Scale Information for Improved Tiny Head Tracking
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_ | 1866908383325454336 |
|---|---|
| author | Kim, Jisu Mattingly, Alex Lee, Eung-Joo Riggan, Benjamin S. |
| author_facet | Kim, Jisu Mattingly, Alex Lee, Eung-Joo Riggan, Benjamin S. |
| contents | Head detection and tracking are essential for downstream tasks, but current methods often require large computational budgets, which increase latencies and ties up resources (e.g., processors, memory, and bandwidth). To address this, we propose a framework to enhance tiny head detection and tracking by optimizing the balance between performance and efficiency. Our framework integrates (1) a cross-domain detection loss, (2) a multi-scale module, and (3) a small receptive field detection mechanism. These innovations enhance detection by bridging the gap between large and small detectors, capturing high-frequency details at multiple scales during training, and using filters with small receptive fields to detect tiny heads. Evaluations on the CroHD and CrowdHuman datasets show improved Multiple Object Tracking Accuracy (MOTA) and mean Average Precision (mAP), demonstrating the effectiveness of our approach in crowded scenes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_22677 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Using Cross-Domain Detection Loss to Infer Multi-Scale Information for Improved Tiny Head Tracking Kim, Jisu Mattingly, Alex Lee, Eung-Joo Riggan, Benjamin S. Computer Vision and Pattern Recognition Head detection and tracking are essential for downstream tasks, but current methods often require large computational budgets, which increase latencies and ties up resources (e.g., processors, memory, and bandwidth). To address this, we propose a framework to enhance tiny head detection and tracking by optimizing the balance between performance and efficiency. Our framework integrates (1) a cross-domain detection loss, (2) a multi-scale module, and (3) a small receptive field detection mechanism. These innovations enhance detection by bridging the gap between large and small detectors, capturing high-frequency details at multiple scales during training, and using filters with small receptive fields to detect tiny heads. Evaluations on the CroHD and CrowdHuman datasets show improved Multiple Object Tracking Accuracy (MOTA) and mean Average Precision (mAP), demonstrating the effectiveness of our approach in crowded scenes. |
| title | Using Cross-Domain Detection Loss to Infer Multi-Scale Information for Improved Tiny Head Tracking |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.22677 |