A new method of image classification based on local appearance and context information

In this paper, we present a new method of image classification based on local appearance and context information. At first, informative or representative local features are selected based on SVM classifier; and then the related visual context information are extracted to keep the robustness to object occlusion and background clutter. Finally, general probabilistic models are built to implement image classification by integrating local invariant characteristics and context information. Experimental results show that the proposed method can outperform other previous methods on several datasets with limited and unnormalized training samples even for large scale classes of objects, therefore the significance of appearance-based discriminative classifiers is demonstrated and confirmed.