2 code implementations • 9 Jul 2018 • Pengchong Jin, Vivek Rathod, Xiangxin Zhu
We share box predictors across all scales, and replace convolution between scales with max pooling.
no code implementations • 25 Nov 2017 • Siyang Li, Xiangxin Zhu, Qin Huang, Hao Xu, C. -C. Jay Kuo
When supervising an object detector with weakly labeled data, most existing approaches are prone to trapping in the discriminative object parts, e. g., finding the face of a cat instead of the full body, due to lacking the supervision on the extent of full objects.
no code implementations • 5 Mar 2015 • Xiangxin Zhu, Carl Vondrick, Charless Fowlkes, Deva Ramanan
Datasets for training object recognition systems are steadily increasing in size.
no code implementations • CVPR 2014 • Xiangxin Zhu, Dragomir Anguelov, Deva Ramanan
We argue that object subcategories follow a long-tail distribution: a few subcategories are common, while many are rare.