计算机科学 ›› 2014, Vol. 41 ›› Issue (12): 260-263.doi: 10.11896/j.issn.1002-137X.2014.12.056
吴丽娜,黄雅平,郑翔
WU Li-na,HUANG Ya-ping and ZHENG Xiang
摘要: 在分类新类别图像时,词袋模型总需要重新学习视觉词典及分类器,而不能充分利用已经学习好的视觉词典。运用迁移学习的思想,提出一种视觉短语的迁移学习算法。这种视觉短语不仅包含图像的局部不变特征,而且包含特征间的空间结构信息,能更有效地描述不同类别图像之间的共同特征。在分类新类别图像时,算法通过迁移视觉短语而不是重新学习视觉词典,来完成图像分类任务。实验结果证明这种迁移算法能有效地利用已有知识,在分类 新类别图像时取得很好的效果,而且还能适用于仅有少量训练样本的图像分类任务。
[1] Yao Y,Doretto G.Boosting for transfer learning with multiple sources[C]∥IEEE Conference on Computer Vision and Pattern Recognition.2010:1855-1862 [2] Lin D,Xing A,Zhang Z J.Double-bootstrapping source data selection for instance-based transfer learning[J].Pattern Recognition Letters,2013,34(11):1279-1285 [3] Raina R,Battle A,Lee H,et al.Self-taught Learning:Transfer Learning from Unlabeled Data[C]∥Proceedings of the 24th international conference on Machine learning.2007:759-766 [4] Long M,Ding G,Wang J.Transfer Sparse Coding for RobustImage Representation[C]∥IEEE Conference on Computer Vision and Pattern Recognition.2013 [5] Perronnin F.Universal and Adapted Vocabularies for GenericVisual Categorization[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(7):1243-1256 [6] 谢尧芳,苏松志,李绍滋.基于稀疏编码的迁移学习及其在行人检测中的应用[J].厦门大学学报:自然科学版,2010,49(2):186-192 (下转第274页)(上接第263页) [7] Lazebnik S,Schmid C,Ponce J.Beyond bags of features:spatial pyramid matching for recognizing natural scene categories[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition.2006:2169-2178 [8] Yang J,Yu K,Gong Y,et al.Linear Spatial Pyramid MatchingUsing Sparse Coding for Image Classification[C]∥IEEE Conference on Computer Vision and Pattern Recognition.2009:1794-1801 [9] Russakovsky O,Lin Y,Yu K,et al.Object-centric spatial pooling for image classification[C]∥European Conference on Computer Vision.2012 [10] Yao Bang-peng,Li Fei-fei.Recognizing Human-Object Interac-tions in Still Images by Modeling the Mutual Context of Objects and Human Poses[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(9):1691-1703 |
No related articles found! |
|