1 / 22

Color Attributes for Object Detection

Color Attributes for Object Detection. CVPR 2012 Poster. Outline. Introduction Color attributes for object detection Coloring object detection Cartoon character detection Experiment Conclusion. Introduction. Introduction. Why with color attributes?

kimball
Download Presentation

Color Attributes for Object Detection

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Color Attributes for Object Detection CVPR 2012 Poster

  2. Outline • Introduction • Color attributes for object detection • Coloring object detection • Cartoon character detection • Experiment • Conclusion

  3. Introduction

  4. Introduction • Why with color attributes? • There are few approach apply color for object detection on a single class. [15] W. R. Schwartz, A. Kembhavi, D. Harwood, and L. S. Davis. Human detection using partial least squares analysis. In ICCV, 2009. [22] S. Walk, N. Majer, K. Schindler, and B. Schiele. New features and insights for pedestrian detection. In CVPR, 2010. [2] R. M. Anwer, D. Vazquez, and A. M. Lopez. Color contribution to part-based person detection in different types of scenarios.

  5. Introduction • Feature combination: Early fusion

  6. Introduction • Feature combination: Late fusion

  7. Introduction • Photometric invariance • provides guidelines on how to ensure invariance with respect to such event • Compactness

  8. Color attributes for object detection • Color descriptor • Robust hue descriptor(HUE): • Opponent derivative descriptor(OPP):

  9. Color attributes for object detection • Color descriptor • Color names(CN):

  10. Color attributes for object detection • Color descriptor evaluation

  11. Color attributes for object detection

  12. Coloring object detection • Coloring part-based object detection

  13. Coloring object detection • Coloring efficient subwindow search • Based on a bag-of-words representation of the image.

  14. Cartoon character detection • New dataset: 586 images , 18 categories

  15. Experiment • On the PASCAL VOC datasets • Coloring part-based object detection

  16. Experiment

  17. Experiment • On the PASCAL VOC 2009

  18. Experiment • On the PASCAL VOC datasets • Coloring ESS-based object detection • Provide good localization results on cat and dog of the PASCAL VOC 2007. • Cat: 20.7%->22.3% • Dog: 13.8%->15.8%

  19. Experiment • On the PASCAL VOC datasets • Comparison with state-of-the-art results

  20. Experiment • On the cartoon dataset • Coloring part-based object detection

  21. Experiment • On the cartoon dataset • Coloring ESS-based object detection

  22. Conclusion • Know that the difference of two feature combination. • Realize that the problem of incorporating color for object detection.

More Related