| Discriminatively trained part based model armed with biased saliency | |
| Yu, Huapeng1,2,3; Chang, Yongxin1,2,3; Lu, Pei1,2,3; Xu, Zhiyong1; Fu, Chengyu1; Wang, Yafei2 | |
| Volume | 9255 |
| Pages | 92553H |
| 2015 | |
| Language | 英语 |
| ISSN | 0277-786X |
| DOI | 10.1117/12.2064960 |
| Indexed By | SCI ; Ei |
| Subtype | 会议论文 |
| Abstract | Discriminatively trained Part based Model (DPM) is one of the state-of-the-art object detectors. However, DPM complies little with real vision procedure. In this paper, we try arming DPM with biologically inspired approaches. On the one hand, we use Gabor instead of Histogram of Oriented Gradient (HOG) as low level features to simulate the receptive fields of simple cells. We show Gabor outperforms or is on par with HOG. On the other hand, we learn biased saliency of the object with the same Gabor features to simulate the search procedure of real vision. We combine DPM and biased saliency in a single Bayesian framework, which at least partially reflects the interactions between top-down and bottom-up vision procedures. We show these biologically inspired procedures can effectively improve the performance and efficiency of DPM. We present experimental results on both challenging PASCAL VOC2007 dataset and publicly available sequences. © 2015 SPIE.; Discriminatively trained Part based Model (DPM) is one of the state-of-the-art object detectors. However, DPM complies little with real vision procedure. In this paper, we try arming DPM with biologically inspired approaches. On the one hand, we use Gabor instead of Histogram of Oriented Gradient (HOG) as low level features to simulate the receptive fields of simple cells. We show Gabor outperforms or is on par with HOG. On the other hand, we learn biased saliency of the object with the same Gabor features to simulate the search procedure of real vision. We combine DPM and biased saliency in a single Bayesian framework, which at least partially reflects the interactions between top-down and bottom-up vision procedures. We show these biologically inspired procedures can effectively improve the performance and efficiency of DPM. We present experimental results on both challenging PASCAL VOC2007 dataset and publicly available sequences. © 2015 SPIE. |
| Conference Name | Proceedings of SPIE: 20th International Symposium on High Power Systems and Applications 2014, HPLS and A 2014 |
| Conference Date | 2015 |
| Citation statistics | |
| Document Type | 会议论文 |
| Identifier | http://ir.ioe.ac.cn/handle/181551/7707 |
| Collection | 光电探测与信号处理研究室（五室） |
| Corresponding Author | Yu, Huapeng |
| Affiliation | 1. Institute of Optics and Electronics, Key Laboratory of Beam Control, Chinese Academy of Sciences, Chengdu, China 2. School of Optoelectronic Information, University of Electronic Science and Technology of China, Chengdu, China 3. University of Chinese Academy of Sciences, Beijing, China |
| Recommended Citation GB/T 7714 | Yu, Huapeng,Chang, Yongxin,Lu, Pei,et al. Discriminatively trained part based model armed with biased saliency[C],2015:92553H. |
| Files in This Item: | ||||||
| File Name/Size | DocType | Version | Access | License | ||
| 2015-2052.pdf（919KB） | 会议论文 | 开放获取 | CC BY-NC-SA | Application Full Text | ||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment