Abstract
Mobile product image search aims at identifying a product, or retrieving similar products from a database based on a photo captured from a mobile phone camera. Application of traditional image retrieval methods (e.g. bag-of-words) to mobile visual search has been shown to be effective in identifying duplicate/near-duplicate photos, near-planar and textured objects such as landmarks, books/cd covers. However, retrieving more general product categories is still a challenging research problem due to variations in viewpoint, illumination, scale, the existence of blur and background clutter in the query image, etc. In this paper, we propose a new approach that can simultaneously extract the product instance from the query, identify the instance, and retrieve visually similar product images. Based on the observation that good query segmentation helps improve retrieval accuracy and good search results provide good priors for segmentation, we formulate our approach in an iterative scheme to improve both query segmentation and retrieval accuracy. To this end, a weighted object mask voting algorithm is proposed based on a spatially-constrained model, which allows robust localization and segmentation of the query object, and achieves significantly better retrieval accuracy than previous methods. We show the effectiveness of our approach by applying it to a large, real-world product image dataset and a new object category dataset.
| Language | English (US) |
|---|---|
| Title of host publication | Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings |
| Pages | 114-127 |
| Number of pages | 14 |
| Edition | PART 4 |
| DOIs | |
| State | Published - Oct 30 2012 |
| Event | 12th European Conference on Computer Vision, ECCV 2012 - Florence, Italy Duration: Oct 7 2012 → Oct 13 2012 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| No. | PART 4 |
| Volume | 7575 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Other
| Other | 12th European Conference on Computer Vision, ECCV 2012 |
|---|---|
| Country | Italy |
| City | Florence |
| Period | 10/7/12 → 10/13/12 |
Fingerprint
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science(all)
Cite this
}
Mobile product image search by automatic query object extraction. / Shen, Xiaohui; Lin, Zhe; Brandt, Jonathan; Wu, Ying.
Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings. PART 4. ed. 2012. p. 114-127 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7575 LNCS, No. PART 4).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Mobile product image search by automatic query object extraction
AU - Shen,Xiaohui
AU - Lin,Zhe
AU - Brandt,Jonathan
AU - Wu,Ying
PY - 2012/10/30
Y1 - 2012/10/30
N2 - Mobile product image search aims at identifying a product, or retrieving similar products from a database based on a photo captured from a mobile phone camera. Application of traditional image retrieval methods (e.g. bag-of-words) to mobile visual search has been shown to be effective in identifying duplicate/near-duplicate photos, near-planar and textured objects such as landmarks, books/cd covers. However, retrieving more general product categories is still a challenging research problem due to variations in viewpoint, illumination, scale, the existence of blur and background clutter in the query image, etc. In this paper, we propose a new approach that can simultaneously extract the product instance from the query, identify the instance, and retrieve visually similar product images. Based on the observation that good query segmentation helps improve retrieval accuracy and good search results provide good priors for segmentation, we formulate our approach in an iterative scheme to improve both query segmentation and retrieval accuracy. To this end, a weighted object mask voting algorithm is proposed based on a spatially-constrained model, which allows robust localization and segmentation of the query object, and achieves significantly better retrieval accuracy than previous methods. We show the effectiveness of our approach by applying it to a large, real-world product image dataset and a new object category dataset.
AB - Mobile product image search aims at identifying a product, or retrieving similar products from a database based on a photo captured from a mobile phone camera. Application of traditional image retrieval methods (e.g. bag-of-words) to mobile visual search has been shown to be effective in identifying duplicate/near-duplicate photos, near-planar and textured objects such as landmarks, books/cd covers. However, retrieving more general product categories is still a challenging research problem due to variations in viewpoint, illumination, scale, the existence of blur and background clutter in the query image, etc. In this paper, we propose a new approach that can simultaneously extract the product instance from the query, identify the instance, and retrieve visually similar product images. Based on the observation that good query segmentation helps improve retrieval accuracy and good search results provide good priors for segmentation, we formulate our approach in an iterative scheme to improve both query segmentation and retrieval accuracy. To this end, a weighted object mask voting algorithm is proposed based on a spatially-constrained model, which allows robust localization and segmentation of the query object, and achieves significantly better retrieval accuracy than previous methods. We show the effectiveness of our approach by applying it to a large, real-world product image dataset and a new object category dataset.
UR - http://www.scopus.com/inward/record.url?scp=84867871718&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84867871718&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-33765-9_9
DO - 10.1007/978-3-642-33765-9_9
M3 - Conference contribution
SN - 9783642337642
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 114
EP - 127
BT - Computer Vision, ECCV 2012 - 12th European Conference on Computer Vision, Proceedings
ER -