基于局部分塊和模型更新的視覺跟蹤算法
doi: 10.11999/JEIT141134
基金項目:
國家自然科學(xué)基金(61175029, 61473309)和陜西省自然科學(xué)基金(2011JM8015)資助課題
Visual Object Tracking Method Based on Local Patch Model and Model Update
-
摘要: 針對目標(biāo)跟蹤過程中的目標(biāo)表觀變化、背景干擾及發(fā)生遮擋等問題，該文提出一種基于局部分塊和模型更新的視覺跟蹤算法。該文采用粗搜索與精搜索相結(jié)合的雙層搜索方法來提高目標(biāo)的定位精度。首先，在包含部分背景區(qū)域的初始跟蹤區(qū)域內(nèi)構(gòu)建目標(biāo)模型。然后，利用基于積分直方圖的局部窮搜索算法初步確定目標(biāo)的位置，接著在當(dāng)前跟蹤區(qū)域內(nèi)通過分塊學(xué)習(xí)來精確搜索目標(biāo)的最終位置。最后，利用創(chuàng)建的模型更新域?qū)δ繕?biāo)模型進行更新。該文主要針對分塊跟蹤中的背景抑制、模型更新等方面進行了研究，實驗結(jié)果表明該算法對目標(biāo)表觀變化、背景干擾及遮擋情況的處理能力都有所增強。
-
關(guān)鍵詞:
- 視覺跟蹤 /
- 局部分塊模型 /
- 窮搜索 /
- 局部分塊學(xué)習(xí) /
- 模型更新
Abstract: In order to solve the problems of appearance change, background distraction and occlusion in the object tracking, an efficient algorithm for visual tracking based on the local patch model and model update is proposed. This paper combines rough-search and precise-search to enhance the tracking precision. Firstly, it constructs the local patch model according to the initialized tracking area which includes some background areas. Secondly, the target is preliminarily located through the local exhaustive search algorithm based on the integral histogram, then the final position of the target is calculated through the local patches learning. Finally, the local patch model is updated with the retained sequence during the tracking process. This paper mainly studies the search strategy, background restraining and model update, and the experimental results show that the proposed method obtains a distinct improvement in coping with appearance change, background distraction and occlusion.-
Key words:
- Visual tracking /
- Local patch model /
- Exhaustive search /
- Local patches learning /
- Model update
-
Yang Han-xuan, Shao Ling, Zheng Feng, et al.. Recent advances and trends in visual tracking: a review[J]. Neurocomputing, 2011, 74(18): 3823-3831. Wu Yi, Lim J, and Yang M H. Online object tracking: a benchmark[C]. Proceedings of the Computer Vision and Pattern Recognition, Portland, United States, 2013: 2411-2418. Smeulders A W M, Chu D M, Cucchiara R, et al.. Visual tracking: an experimental survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013: DOI: 10. 1109/TPAMI. 2013. 230. Lu Zhang and Laurens V D M. Structure preserving object tracking[C]. Proceedings of the Computer Vision and Pattern Recognition, Portland, United States, 2013: 1838-1845. Comaniciu D, Ramesh V, and Meer P. Real-time tracking of non-rigid objects using mean shift[C]. Proceedings of the Computer Vision and Pattern Recognition, Hilton Head Island, United States, 2000: 142-149. Comaniciu D, Ramesh V, and Meer P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577. Babenko B, Yang M H, and Belongie S. Robust object tracking with online multiple instance learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8): 1619-1632. Wang Dong, Lu Hu-chuan, and Yang M H. Online object tracking with sparse prototypes[J]. IEEE Transactions on Image Processing, 2013, 22(1): 314-315. Adam A, Rivlin E, and Shimshoni I. Robust fragments-based tracking using the integral histogram[C]. Proceedings of the Computer Vision and Pattern Recognition, New York, United States, 2006: 798-805. Nejhum S, Ho J, and Yang M H. Online visual tracking with histograms and articulating blocks[J]. Computer Vision and Image Understanding, 2010, 114(8): 901-914. 董文會, 常發(fā)亮, 李天平. 融合顏色直方圖及SIFT特征的自適應(yīng)分塊目標(biāo)跟蹤方法[J]. 電子與信息學(xué)報, 2013, 35(4): 770-776. Dong Wen-hui, Chang Fa-liang, and Li Tian-ping. Adaptive fragments-based target tracking method fusing color histogram and SIFT features[J]. Journal of Electronics Information Technology, 2013, 35(4): 770-776. Wang Shu, Lu Hu-chuan, Yang Fan, et al.. Superpixel tracking[C]. Proceedings of the IEEE International Conference on Computer Vision, Barcelona, Spain, 2011: 1323-1330. Yang Fan, Lu Hu-chuan, and Yang M H. Robust superpixel tracking[J]. IEEE Transactions on Image Processing, 2014, 23(4): 1639-1651. Matthews I, Ishikawa T, and Baker S. The template update problem[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(6): 810-815. Porkili F. Integral histogram: a fast way to extract histograms in cartesian spaces[C]. Proceedings of the Computer Vision and Pattern Recognition, San Diego, United States, 2005: 829-836. -
計量
- 文章訪問數(shù): 1442
- HTML全文瀏覽量: 204
- PDF下載量: 368
- 被引次數(shù): 0
下載:
