End-to-End Trained Sparse Coding Network with Spatial Pyramid Pooling for Image Classification

2019 ◽  
Vol 50 (3) ◽  
pp. 2021-2036 ◽  
Author(s):  
Boheng Chen ◽  
Yige Wang ◽  
Gang Wei ◽  
Jie Li ◽  
Biyun Ma
Keyword(s):  
Image Classification ◽  
Sparse Coding ◽  
Spatial Pyramid Pooling ◽  
End To End ◽  
Spatial Pyramid

Image classification using spatial pyramid robust sparse coding

2013 ◽  
Vol 34 (9) ◽  
pp. 1046-1052 ◽  
Author(s):  
Chunjie Zhang ◽  
Shuhui Wang ◽  
Qingming Huang ◽  
Jing Liu ◽  
Chao Liang ◽  
...  
Keyword(s):  
Image Classification ◽  
Sparse Coding ◽  
Spatial Pyramid

Hierarchical spatial pyramid max pooling based on SIFT features and sparse coding for image classification

2013 ◽  
Vol 7 (2) ◽  
pp. 144-150 ◽  
Author(s):  
Hong Han ◽  
Qiqiang Han ◽  
Xiaojun Li ◽  
Jianyin Gu
Keyword(s):  
Image Classification ◽  
Sparse Coding ◽  
Max Pooling ◽  
Sift Features ◽  
Spatial Pyramid

scholarly journals A Fast Sparse Coding Method for Image Classification

2019 ◽  
Vol 9 (3) ◽  
pp. 505 ◽  
Author(s):  
Mujun Zang ◽  
Dunwei Wen ◽  
Tong Liu ◽  
Hailin Zou ◽  
Chanjuan Liu
Keyword(s):  
Computational Complexity ◽  
Image Classification ◽  
Sparse Coding ◽  
Cluster Method ◽  
Spatial Pyramid Matching ◽  
Mutual Dependence ◽  
Coding Method ◽  
Pyramid Matching ◽  
Spatial Pyramid ◽  
High Computational Complexity

Image classification is an important problem in computer vision. The sparse coding spatial pyramid matching (ScSPM) framework is widely used in this field. However, the sparse coding cannot effectively handle very large training sets because of its high computational complexity, and ignoring the mutual dependence among local features results in highly variable sparse codes even for similar features. To overcome the shortcomings of previous sparse coding algorithm, we present an image classification method, which replaces the sparse dictionary with a stable dictionary learned via low computational complexity clustering, more specifically, a k-medoids cluster method optimized by k-means++. The proposed method can reduce the learning complexity and improve the feature’s stability. In the experiments, we compared the effectiveness of our method with the existing ScSPM method and its improved versions. We evaluated our approach on two diverse datasets: Caltech-101 and UIUC-Sports. The results show that our method can increase the accuracy of spatial pyramid matching, which suggests that our method is capable of improving performance of sparse coding features.

Sign in / Sign up

Export Citation Format

Share Document