Randomized spatial partition for scene recognition
The spatial layout of images plays a critical role in natural scene analysis. Despite previous work, e.g., spatial pyramid matching, how to design optimal spatial layout for scene classification remains an open problem due to the large variations of scene categories. This paper presents a novel imag...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Conference Paper |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/100601 http://hdl.handle.net/10220/17891 |
| _version_ | 1826109829176885248 |
|---|---|
| author | Jiang, Yuning Yuan, Junsong Yu, Gang |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Jiang, Yuning Yuan, Junsong Yu, Gang |
| author_sort | Jiang, Yuning |
| collection | NTU |
| description | The spatial layout of images plays a critical role in natural scene analysis. Despite previous work, e.g., spatial pyramid matching, how to design optimal spatial layout for scene classification remains an open problem due to the large variations of scene categories. This paper presents a novel image representation method, with the objective to characterize the image layout by various patterns, in the form of randomized spatial partition (RSP). The RSP-based
image representation makes it possible to mine the most descriptive image layout pattern for each category of scenes, and then combine them by training a discriminative classifier, i.e., the proposed ORSP classifier. Besides RSP image
representation, another powerful classifier, called the BRSP classifier, is also proposed. By weighting a sequence of various partition patterns via boosting, the BRSP classifier is more robust to the intra-class variations hence leads to a more
accurate classification. Both RSP-based classifiers are tested on three publicly available scene datasets. The experimental results highlight the effectiveness of the proposed methods. |
| first_indexed | 2024-10-01T02:24:29Z |
| format | Conference Paper |
| id | ntu-10356/100601 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:24:29Z |
| publishDate | 2013 |
| record_format | dspace |
| spelling | ntu-10356/1006012020-03-07T13:24:50Z Randomized spatial partition for scene recognition Jiang, Yuning Yuan, Junsong Yu, Gang School of Electrical and Electronic Engineering European conference on Computer Vision (12th : 2012 : Florence, Italy) DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition The spatial layout of images plays a critical role in natural scene analysis. Despite previous work, e.g., spatial pyramid matching, how to design optimal spatial layout for scene classification remains an open problem due to the large variations of scene categories. This paper presents a novel image representation method, with the objective to characterize the image layout by various patterns, in the form of randomized spatial partition (RSP). The RSP-based image representation makes it possible to mine the most descriptive image layout pattern for each category of scenes, and then combine them by training a discriminative classifier, i.e., the proposed ORSP classifier. Besides RSP image representation, another powerful classifier, called the BRSP classifier, is also proposed. By weighting a sequence of various partition patterns via boosting, the BRSP classifier is more robust to the intra-class variations hence leads to a more accurate classification. Both RSP-based classifiers are tested on three publicly available scene datasets. The experimental results highlight the effectiveness of the proposed methods. Accepted version 2013-11-29T01:18:39Z 2019-12-06T20:25:12Z 2013-11-29T01:18:39Z 2019-12-06T20:25:12Z 2012 2012 Conference Paper Jiang, Y., Yuan, J., & Yu, G. (2012). Randomized Spatial Partition for Scene Recognition. Proceedings of the 12th European conference on Computer Vision (ECCV12), pp.730-743. https://hdl.handle.net/10356/100601 http://hdl.handle.net/10220/17891 10.1007/978-3-642-33709-3_52 en © 2012 Springer-Verlag Berlin Heidelberg. This is the author created version of a work that has been peer reviewed and accepted for publication by Proceedings of the 12th European conference on Computer Vision (ECCV12), Springer-Verlag Berlin Heidelberg. It incorporates referee’s comments but changes resulting from the publishing process, such as copyediting, structural formatting, may not be reflected in this document. The published version is available at: [http://dx.doi.org/10.1007/978-3-642-33709-3_52]. 14 p. application/pdf |
| spellingShingle | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition Jiang, Yuning Yuan, Junsong Yu, Gang Randomized spatial partition for scene recognition |
| title | Randomized spatial partition for scene recognition |
| title_full | Randomized spatial partition for scene recognition |
| title_fullStr | Randomized spatial partition for scene recognition |
| title_full_unstemmed | Randomized spatial partition for scene recognition |
| title_short | Randomized spatial partition for scene recognition |
| title_sort | randomized spatial partition for scene recognition |
| topic | DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition |
| url | https://hdl.handle.net/10356/100601 http://hdl.handle.net/10220/17891 |
| work_keys_str_mv | AT jiangyuning randomizedspatialpartitionforscenerecognition AT yuanjunsong randomizedspatialpartitionforscenerecognition AT yugang randomizedspatialpartitionforscenerecognition |