Name
Abstract | ||
|---|---|---|
Semantic understanding of images remains an important research challenge for the image and video retrieval community. A novel natural scene retrieval method based on non-negative sparse coding is proposed in this paper. It firstly combines non-negative sparse coding with spatial pyramid matching for feature extraction and representation. Then, based on sparse coding, it ranks the Euclidean distances from the query image to each of the K-nearest neighbors in database. With the help of SIFT flow and label transfer, we finally realize the segmentation and recognition for the query images. The experimental results show that the proposed method has higher relevant relationship between the query image and each of the K-nearest neighbors in database than the scene retrieval method based on GIST. And the good performances of our method will be greatly helpful for the following image understanding. |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1109/CICSyN.2012.60 | CICSyN |
Keywords | Field | DocType |
novel natural scene retrieval,query image,non-negative sparse coding,natural scene retrieval,sparse coding,non-negative sparse,scene retrieval method,following image understanding,video retrieval community,k-nearest neighbor,computational geometry,encoding,image segmentation,feature extraction,computational modeling,visualization,k nearest neighbors,semantics,computer vision,scale invariant feature transform | Computer vision,Scale-invariant feature transform,Pattern recognition,Segmentation,Computer science,Visualization,Neural coding,Image segmentation,Feature extraction,Pyramid,Artificial intelligence,Encoding (memory) | Conference |
Citations | PageRank | References |
0 | 0.34 | 9 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Min Wang | 1 | 62 | 4.30 |
| Xiao-hui Yang | 2 | 0 | 0.34 |
| Lixin Han | 3 | 135 | 14.47 |
| Rong Chu | 4 | 5 | 1.80 |