| DeepBE: Learning Deep Binary Encoding for Multi-Label Classification | |
Li, Chenghua1,2 ; Kang, Qi4; Ge, Guojing1; Song, Qiang1,2; Lu, Hanqing1,2; Cheng, Jian1,2,3
| |
| 2016 | |
| Conference Name | IEEE Conference on Computer Vision and Pattern Recognition Workshop |
| Conference Date | 2016.6.27-6.30 |
| Conference Place | Las Vegas, NV, USA |
| Abstract | The track 2 and track 3 of ChaLearn 2016 can be considered as Multi-Label Classification problems. We present a framework of learning deep binary encoding (DeepBE) to deal with multi-label problems by transforming multi-labels to single labels. The transformation of DeepBE is in a hidden pattern, which can be well addressed by deep convolutions neural networks (CNNs). Furthermore, we adopt an ensemble strategy to enhance the learning robustness. This strategy is inspired by its effectiveness in fine-grained image recognition (FGIR) problem, while most of face related tasks such as track 2 and track 3 are also FGIR problems. By DeepBE, we got 5.45% and 10.84% mean square error for track 2 and track 3 respectively. Additionally, we proposed an algorithm adaption method to treat the multiplelabels of track 2 directly and got 6.84% mean square error. |
| Keyword | Chalearn2016 Deepbe Multi-label Classification |
| Indexed By | EI |
| Document Type | 会议论文 |
| Identifier | http://ir.ia.ac.cn/handle/173211/20140 |
| Collection | 模式识别国家重点实验室_图像与视频分析 |
| Affiliation | 1.Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China 2.University of Chinese Academy of Sciences 3.Research Center for Brain-inspired Intelligence, Institute of Automation, Chinese Academy of Sciences 4.Beijing Institute Of Technology |
| Recommended Citation GB/T 7714 | Li, Chenghua,Kang, Qi,Ge, Guojing,et al. DeepBE: Learning Deep Binary Encoding for Multi-Label Classification[C],2016. |
| Files in This Item: | Download All | |||||
| File Name/Size | DocType | Version | Access | License | ||
| DeepBE Learning Deep（3165KB） | 会议论文 | 开放获取 | CC BY-NC-SA | View Download | ||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment