Advanced search
2 files | 1.35 MB Add to list
Author
Organization
Project
Abstract
Occlusion is one of the most challenging problems in single-view pedestrian detection. To alleviate the occlusion problem, multi-view systems have been exploited to fully acquire and recognize blocked targets. Most often, methods from the literature exploit perspective transformation to aggregate different sensing view angles of the scene, but projection distortion issues cause spatial structure break and prevent these methods from fully exploring the projected features. In this paper, we propose a novel approach, Multi-view Target Transformation (MVTT), to address the distortion problem inherent in multi-view aggregation by encoding the full target features and limiting the area of interest of the projected features. Experiment results show that the performance of our proposed method compares favorably against recent relevant methods on public datasets. The ablation studies also confirm the effectiveness of the proposed components.
Keywords
TRACKING

Downloads

  • (...).pdf
    • full text (Accepted manuscript)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.18 MB
  • (...).pdf
    • supplementary material
    • |
    • UGent only
    • |
    • PDF
    • |
    • 177.81 KB

Citation

Please use this url to cite or link to this publication:

MLA
Lee, Wei-Yu, et al. “Multi-View Target Transformation for Pedestrian Detection.” 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), IEEE, 2023, pp. 90–99, doi:10.1109/WACVW58289.2023.00014.
APA
Lee, W.-Y., Jovanov, L., & Philips, W. (2023). Multi-view target transformation for pedestrian detection. 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), 90–99. https://doi.org/10.1109/WACVW58289.2023.00014
Chicago author-date
Lee, Wei-Yu, Ljubomir Jovanov, and Wilfried Philips. 2023. “Multi-View Target Transformation for Pedestrian Detection.” In 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), 90–99. IEEE. https://doi.org/10.1109/WACVW58289.2023.00014.
Chicago author-date (all authors)
Lee, Wei-Yu, Ljubomir Jovanov, and Wilfried Philips. 2023. “Multi-View Target Transformation for Pedestrian Detection.” In 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), 90–99. IEEE. doi:10.1109/WACVW58289.2023.00014.
Vancouver
1.
Lee W-Y, Jovanov L, Philips W. Multi-view target transformation for pedestrian detection. In: 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW). IEEE; 2023. p. 90–9.
IEEE
[1]
W.-Y. Lee, L. Jovanov, and W. Philips, “Multi-view target transformation for pedestrian detection,” in 2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), Hawaii, USA, 2023, pp. 90–99.
@inproceedings{01GPFHH4K0F79ADJ3BJCX976MV,
  abstract     = {{Occlusion is one of the most challenging problems in single-view pedestrian detection. To alleviate the occlusion problem, multi-view systems have been exploited to fully acquire and recognize blocked targets. Most often, methods from the literature exploit perspective transformation to aggregate different sensing view angles of the scene, but projection distortion issues cause spatial structure break and prevent these methods from fully exploring the projected features. In this paper, we propose a novel approach, Multi-view Target Transformation (MVTT), to address the distortion problem inherent in multi-view aggregation by encoding the full target features and limiting the area of interest of the projected features. Experiment results show that the performance of our proposed method compares favorably against recent relevant methods on public datasets. The ablation studies also confirm the effectiveness of the proposed components.}},
  author       = {{Lee, Wei-Yu and Jovanov, Ljubomir and Philips, Wilfried}},
  booktitle    = {{2023 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)}},
  isbn         = {{9798350320565}},
  issn         = {{2572-4398}},
  keywords     = {{TRACKING}},
  language     = {{eng}},
  location     = {{Hawaii, USA}},
  pages        = {{90--99}},
  publisher    = {{IEEE}},
  title        = {{Multi-view target transformation for pedestrian detection}},
  url          = {{http://doi.org/10.1109/WACVW58289.2023.00014}},
  year         = {{2023}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: