Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.19032 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910920299511808 |
|---|---|
| author | Ahmad, Niaz Lee, Youngmoon Wang, Guanghui |
| author_facet | Ahmad, Niaz Lee, Youngmoon Wang, Guanghui |
| contents | We introduce VISUALCENT, a unified human pose and instance segmentation framework to address generalizability and scalability limitations to multi person visual human analysis. VISUALCENT leverages centroid based bottom up keypoint detection paradigm and uses Keypoint Heatmap incorporating Disk Representation and KeyCentroid to identify the optimal keypoint coordinates. For the unified segmentation task, an explicit keypoint is defined as a dynamic centroid called MaskCentroid to swiftly cluster pixels to specific human instance during rapid changes in human body movement or significantly occluded environment. Experimental results on COCO and OCHuman datasets demonstrate VISUALCENTs accuracy and real time performance advantages, outperforming existing methods in mAP scores and execution frame rate per second. The implementation is available on the project page. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_19032 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VISUALCENT: Visual Human Analysis using Dynamic Centroid Representation Ahmad, Niaz Lee, Youngmoon Wang, Guanghui Computer Vision and Pattern Recognition Artificial Intelligence We introduce VISUALCENT, a unified human pose and instance segmentation framework to address generalizability and scalability limitations to multi person visual human analysis. VISUALCENT leverages centroid based bottom up keypoint detection paradigm and uses Keypoint Heatmap incorporating Disk Representation and KeyCentroid to identify the optimal keypoint coordinates. For the unified segmentation task, an explicit keypoint is defined as a dynamic centroid called MaskCentroid to swiftly cluster pixels to specific human instance during rapid changes in human body movement or significantly occluded environment. Experimental results on COCO and OCHuman datasets demonstrate VISUALCENTs accuracy and real time performance advantages, outperforming existing methods in mAP scores and execution frame rate per second. The implementation is available on the project page. |
| title | VISUALCENT: Visual Human Analysis using Dynamic Centroid Representation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2504.19032 |