Iterative localization refinement in convolutional neural networks for improved object detection

Kai-Wen Cheng, Yie-Tarng Chen, Wen-Hsien FangICIP 2016 · OpenReview

Abstract

Accurate region proposals are of importance to facilitate object localization in the existing convolutional neural network (CNN)-based object detection methods. This paper presents a novel iterative localization refinement (ILR) method which, undertaken at a mid-layer of a CNN architecture, iteratively refines region proposals in order to match as much ground-truth as possible. The search for the desired bounding box in each iteration is first formulated as a statistical hypothesis testing problem and then solved by a divide-and-conquer paradigm. The proposed ILR is not only data-driven, free of learning, but also compatible with a variety of CNNs. Furthermore, to reduce complexity, an approximate variant based on a refined sampling strategy using linear interpolation is addressed. Simulations show that the proposed method improves the main state-of-the-art works on the PASCAL VOC 2007 dataset.
BibTeX
@inproceedings{cheng2016iterative,
  author = {Kai-Wen Cheng and Yie-Tarng Chen and Wen-Hsien Fang},
  title = {{Iterative localization refinement in convolutional neural networks for improved object detection}},
  booktitle = {ICIP},
  year = {2016},
  url = {https://openreview.net/forum?id=mUi7hk9NCJ},
}

Loading related content...

Figure