Regional Gating Neural Networks for Multi-label Image Classification

Rui-Wei Zhao, Jianguo Li, Yurong Chen, Jia-Ming Liu, Yu-Gang Jiang and Xiangyang Xue

Abstract

This paper proposes a novel deep learning framework for multi-label image classification, namely regional gating neural networks (RGNN). The motivation is two folds. First, global image features (including CNN based features) ignore the underlying context information among different objects in an image. Consequently, people attempt to use information from objectness regions. However, current objectness region proposal algorithms usually produce several thousand region candidates, including many classification irrelevant or even noisy regions. This leads to the second problem: how to select useful contextual regions for image classification. RGNN is an end-to-end deep learning framework that can automatically select contextual region features with specially designed gate units, which are then fused for classification. Because the gate units and the classifier are integrated in the same deep neural network pipeline, we can learn parameters of the network simultaneously. We evaluate the proposed method on PASCAL VOC 2007{\slash}2012 and MS-COCO benchmarks, and results show that RGNN is superior to existing state-of-the-art methods.

Session

Recognition and Physics-based vision

Files

Extended Abstract (PDF, 45K)
Paper (PDF, 587K)

DOI

10.5244/C.30.72
https://dx.doi.org/10.5244/C.30.72

Citation

Bibtex

        @inproceedings{BMVC2016_72,
        	title={Regional Gating Neural Networks for Multi-label Image Classification},
        	author={Rui-Wei Zhao, Jianguo Li, Yurong Chen, Jia-Ming Liu, Yu-Gang Jiang and Xiangyang Xue},
        	year={2016},
        	month={September},
        	pages={72.1-72.12},
        	articleno={72},
        	numpages={12},
        	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
        	publisher={BMVA Press},
        	editor={Richard C. Wilson, Edwin R. Hancock and William A. P. Smith},
        	doi={10.5244/C.30.72},
        	isbn={1-901725-59-6},
        	url={https://dx.doi.org/10.5244/C.30.72}
        }