Learning Region Features for Object Detection

Jiayuan Gu, Han Hu, Liwei Wang, Yichen Wei, Jifeng Dai; Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 381-395

Abstract


While most steps in the modern object detection methods are learnable, the region feature extraction step remains largely hand-crafted, featured by RoI pooling methods. This work proposes a general viewpoint that unifies existing region feature extraction methods and a novel method that is end-to-end learnable. The proposed method removes most heuristic choices and outperforms its RoI pooling counterparts. It moves further towards emph{fully learnable object detection}.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{Gu_2018_ECCV,
author = {Gu, Jiayuan and Hu, Han and Wang, Liwei and Wei, Yichen and Dai, Jifeng},
title = {Learning Region Features for Object Detection},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
month = {September},
year = {2018}
}