Pose Pooling Kernels for Sub-Category Recognition
Title | Pose Pooling Kernels for Sub-Category Recognition |
Publication Type | Conference Paper |
Year of Publication | 2012 |
Authors | Zhang, N., Farrell R., & Darrell T. |
Page(s) | 3665-3672 |
Other Numbers | 3300 |
Abstract | The ability to normalize pose based on super-category landmarks can significantly improve models of individual categories when training data are limited. Previous methods have considered the use of volumetric or morphable models for faces and for certain classes of articulated objects. We consider methods which impose fewer representational assumptions on categories of interest, and exploit contemporary detection schemes which consider the ensemble of responses of detectors trained for specific posekeypoint configurations. We develop representations for poselet-based pose normalization using both explicit warping and implicit pooling as mechanisms. Our method defines a pose normalized similarity or kernel function that is suitable for nearest-neighbor or kernel-based learning methods. |
Acknowledgment | This work was partially supported by funding provided to ICSI by the U.S. Defense Advanced Research Projects Agency (DARPA), by the National Science Foundation through awards IIS: 0905647 ("Computer Vision and Online Communities: A Symbiosis") and IIS: 0819984 ("Perceptually Situated Human Robot Dialog "), by Toyota, and by Google. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors or originators and do not necessarily reflect the views of DARPA, of the National Science Foundation, or of the U.S. Government. |
URL | http://www.icsi.berkeley.edu/pubs/vision/ICSI_posepooling12.pdf |
Bibliographic Notes | Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2012), Providence, Rhode Island, pp. 3665-3672 |
Abbreviated Authors | N. Zhang, R. Farrell, and T. Darrell |
ICSI Research Group | Vision |
ICSI Publication Type | Article in conference proceedings |