PS:mproving Object Detection With Deep Convolutional Networks via Bayesian Optimization..___CVPR2015

Improving Object Detection With Deep Convolutional Networks via Bayesian Optimization and Structured Prediction [full paper] (http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Zhang_Improving_Object_Detection_2015_CVPR_paper.pdf)
Yuting Zhang, Kihyuk Sohn, Ruben Villegas, Gang Pan, Honglak Lee

要说的都在摘要里了，localization问题今年好火爆，这篇文章两个点，一个是采用bayesian optimization来产生candidate bounding box，还有一个是structure loss。传统的方法都有类似工作，这篇paper将其扩展到CNN中，具体细节看论文吧。

While the features learned by these high-capacity neural
networks are discriminative for categorization, inaccurate
localization is still a major source of error for detection.
Building upon high-capacity CNN architectures, we ad-
dress the localization problem by 1) using a search algo-
rithm based on Bayesian optimization that sequentially pro-
poses candidate regions for an object bounding box, and
2) training the CNN with a structured loss that explicitly
penalizes the localization inaccuracy.