基于時(shí)序特性的自適應(yīng)增量主成分分析的視覺跟蹤
doi: 10.11999/JEIT141646 cstr: 32379.14.JEIT141646
基金項(xiàng)目:
國家自然科學(xué)基金重大研究計(jì)劃(90820302)和國家自然科學(xué)基金(61175064, 61403426, 61403423)
Adaptive Incremental Principal Component Analysis Visual Tracking Method Based on Temporal Characteristics
Funds:
The Major Research Project of the National Natural Science Foundation of China (90820302)
-
摘要: 當(dāng)前基于增量主成分分析(PCA)學(xué)習(xí)的跟蹤方法存在兩個(gè)問題，首先，觀測模型沒有考慮目標(biāo)外觀變化的連續(xù)性；其次，當(dāng)目標(biāo)外觀的低維流行分布為非線性結(jié)構(gòu)時(shí)，基于固定頻率更新模型的增量PCA學(xué)習(xí)不能適應(yīng)子空間模型的變化。為此，該文首先基于目標(biāo)外觀變化的連續(xù)性，在子空間模型中提出更合理的目標(biāo)先驗(yàn)概率分布假設(shè)。然后，根據(jù)當(dāng)前跟蹤結(jié)果與子空間模型之間的匹配程度，自適應(yīng)調(diào)整遺忘比例因子，使得子空間模型更能適應(yīng)目標(biāo)外觀變化。實(shí)驗(yàn)結(jié)果驗(yàn)證了所提方法能有效提高跟蹤的魯棒性和精度。
-
關(guān)鍵詞:
- 視覺跟蹤 /
- 主成分分析 /
- 增量子空間學(xué)習(xí) /
- 遺忘因子 /
- 自適應(yīng)增量
Abstract: Existing visual tracking methods based on incremental Principal Component Analysis (PCA) learning have two problems. First, the measurement model does not consider the continuation characteristics of the object appearance changes. Second, when the manifold distribution of target appearance is non-linear structure, the incremental principal component analysis learning based on fixed update frequency can not adapt to changes of subspace model. Therefore, the more reasonable a priori probability distribution of targets is proposed based on the continuity of the object appearance changes in the subspace model. Then, according to the matching degree between the current tracking results and the subspace model, the proposed method adaptively adjusts forgetting factor, in order to make the subspace model more adaptable to the object appearance change. Experimental results show that the proposed method can improve the tracking accuracy and robustness. -
Ross D A, Lim J W, Lin R S, et al.. Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2008, 77(1-3): 125-141. Bao C L, Wu Y, Linh H B, et al.. Real time robust L1 tracker using accelerated proximal gradient approach[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Rhode Island, 2012: 1830-1837. MeI X and Ling H B. Robust visual tracking and vehicle classification via sparse representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(11): 2259-2272. Babenko B, Yang M H, Belongie S, et al.. Robust object tracking with online multiple instance learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8): 1619-1632. Grabner H and Bischof H. On-line boosting and vision[C]. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, 2006, 1: 260-267. Avidan S. Ensemble tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(2): 261-271. Yang M, Zhang C X, Wu Y W, et al.. Robust object tracking via online multiple instance metric learning[C]. Electronic Proceedings of the 2013 IEEE International Conference on Multimedia and Expo Workshops, San Jose, 2013: 1-4. Zhong Wei, Lu Hu-chuan, and Yang M. Robust object tracking via sparse collaborative appearance model[J]. IEEE Transactions on Image Processing, 2014, 23(5): 2356-2368. 呂卓紋, 王科俊, 李宏宇, 等. 融合Camshift的在線Adaboost目標(biāo)跟蹤算法[J].?中南大學(xué)學(xué)報(bào)(自然科學(xué)版), 2013, 44(2): 232-238. Lu Zhuo-wen, Wang Ke-jun, Li Hong-yu, et al.. Online Adaboost target tracking algorithm combined fused with Camshift[J]. Journal of Central South University(Science and Technology), 2013, 44(2): 232-238. 錢誠, 張三元. 適用于目標(biāo)跟蹤的加權(quán)增量子空間學(xué)習(xí)算法[J]. 浙江大學(xué)學(xué)報(bào)(工學(xué)版), 2011, 45(12): 2240-2246. Qian Cheng and Zhang San-yuan. Weighted incremental subspace learning algorithm suitable for object tracking[J]. Journal of Zhejiang University (Engineering Science), 2011, 45(12): 2240-2246. Cruz-Mota J, Bierlaire M, and Thiran J. Sample and pixel weighting strategies for robust incremental visual tracking[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2013, 23(5): 898-911. Xie Yuan, Zhang Wen-sheng, Qu Yan-yun, et al.. Discriminative subspace learning with sparse representation view-based model for robust visual tracking[J]. Pattern Recognition, 2014, 47(3): 1383-1394. Ji Zhang-jian, Wang Wei-qiang, and Xu Ning. Robust object tracking via incremental subspace dynamic sparse model[C]. Proceedings of IEEE International Conference on Multimedia and Expo, Chengdu, 2014: 1-6. Guo Yan-wen, Chen Ye, and Tang Feng. Object tracking using learned feature manifolds[J]. Computer Vision and Image Understanding, 2014, 118(1): 128-139. Chen Wei-hua, Cao Li-jun, and Zhang Jun-ge. An adaptive combination of multiple features for robust tracking in real scene[C]. Proceedings of the IEEE International Conference on Computer Vision, Sydney, 2013: 129-136. Yang Han-xuan, Song Zhan, and Chen Ru-nen. An incremental PCA-HOG descriptor for robust visual hand tracking[J]. Lecture Notes in Computer Science, 2010, 6553: 687-695. Moghaddam B and Pentland A. Probabilistic visual learning for object detection[C]. Proceedings of the IEEE International Conference on Computer Vision, Cambridge, 1995: 786-793. Chen Feng, Wang Qing, Wang Song, et al.. Object tracking via appearance modeling and sparse representation[J]. Image and Vision Computing, 2011, 29(11): 787-796. Wang Dong, Lu Hu-chuan, and Yang M H . Online object tracking with sparse prototypes[J]. IEEE Transactions on Image Processing, 2013, 22(1): 314-325. Wang Qing, Chen Feng, Xu Wen-li, et al.. Object tracking via partial least squares analysis[J]. IEEE Transactions on Image Processing, 2012, 21(10): 4454-4465. -
計(jì)量
- 文章訪問數(shù): 1460
- HTML全文瀏覽量: 125
- PDF下載量: 1077
- 被引次數(shù): 0
下載:
