Bundling Features for Large Scale Partial-Duplicate Web Image Search
Download as PPTX, PDF1 like951 views
This document summarizes a research paper about using bundled features for partial-duplicate image retrieval from large datasets. The authors propose representing images using bundled features, where SIFT features are grouped based on their proximity to maximally stable extremal regions. Each bundled feature encodes the visual words of its SIFT features and their relative ordering. This representation allows for efficient geometric verification between bundled features. The authors evaluate their approach on a dataset of 1 million images, finding that using bundled features improves retrieval accuracy over bag-of-words by up to 49% and is more robust to changes in images.
![ﻪ K-D tree
ﻩ pointList = [(2,3), (5,4), (9,6), (4,7), (8,1),
(7,2)]
13](https://image.slidesharecdn.com/bundlefeature-130818050810-phpapp02/85/Bundling-Features-for-Large-Scale-Partial-Duplicate-Web-Image-Search-13-320.jpg)
![Evaluation
ﻪ Baseline
ﻩ Bag-of-features approach with soft
assignment[13]
[13] J. Philbin, O. Chum, M. Isard, J. Sivic, and A. Zisserman.
Lost in quantization: Improving particular object retrieval in large scale image databases.
In CVPR, 2008.
23](https://image.slidesharecdn.com/bundlefeature-130818050810-phpapp02/85/Bundling-Features-for-Large-Scale-Partial-Duplicate-Web-Image-Search-23-320.jpg)
![ﻪ Compare(HE)
ﻩ enhance the with hamming embedding [3] by
adding a 24-bit hamming code to filter out
target features.
[3] H. Jegou, M. Douze, and C. Schmid.
Hamming embedding and weak geometric consistency for large scale image search.
In ECCV, 2008. 24](https://image.slidesharecdn.com/bundlefeature-130818050810-phpapp02/85/Bundling-Features-for-Large-Scale-Partial-Duplicate-Web-Image-Search-24-320.jpg)
Bundling Features for Large Scale Partial-Duplicate Web Image Search
- 1. Bundling Features for Large Scale Partial-Duplicate Web Image Search Zhong Wu∗, Qifa Ke, Michael Isard, and Jian Sun CVPR 2009 Citations: 163
- 2. Outline Introduction Bundled features Image Retrieval using bundled feature Experiments and results Conclusion 2
- 4. Target ﻪ Given a query image, is to locate its near- and partial-duplicate images in a large corpus of web images. 4
- 5. Novel Scheme ﻪ Each group of bundled features becomes much more discriminative than a single feature ﻪ within each group simple and robust geometric constraints can be efficiently enforced. 5
- 7. Related Work ﻪ SIFT(Scale Invariant Feature Transform) ﻩ keypoint & descriptor from the region centered at the keypoint ﻪ MSER(Maximally Stable Extremal Region) ﻩ Affine-covariant stable region + SIFT from the region 7
- 8. 8
- 9. Bundle Features ﻪ SIFT features: S = {sj} ﻪ MSER detections: R = {ri} ﻪ Define bundled feature B = {bi} : bi = {sj|sj ∝ ri, sj ∈ S} ﻪ We discard any MSER detection whose ellipse spans more than half the width or height of the image 9
- 10. 10
- 11. IMAGE RETRIEVAL USING BUNDLED FEATURE 11
- 12. Feature quantization ﻪ Hierarchical k-means ﻩ One million visual words from 50K training images 12
- 13. ﻪ K-D tree ﻩ pointList = [(2,3), (5,4), (9,6), (4,7), (8,1), (7,2)] 13
- 14. Matching bundled features Let p = {pi} and q = {qj} be two bundled features with quantized visual words pi, qj ∈ W ﻪ Define a matching score : ﻪ M(q; p) = Mm(q; p) + λMg(q; p) ﻩ where λ is a weighting parameter 14
- 15. ﻪ Membership term: ﻩ We simply use the number of common visual words between two bundled features to define the membership term Mm(q; p) ﻪ Mm(q; p) = |{pi}| 15
- 16. ﻪ Geometric term: ﻩ Our geometric term performs a weak geometric verification between two bundled features p and q using relative ordering: ﻪ Indicator Function 16
- 17. 2 17
- 18. Indexing and retrieval ﻪ avoids storing and comparing high dimensional local descriptors ﻪ reduces the number of candidate images 18
- 19. Indexing and retrieval ﻪ Voting ﻪ 19
- 20. Indexing and retrieval ﻪ tf ﻩ 100 vocabularies in a document, ‘a’ 3 times ﻩ 0.03 (3/100) ﻪ idf ﻩ 1,000 documents have ‘a’, total number of documents 10,000,000 ﻩ 9.21 ( ln(10,000,000 / 1,000) ) ﻪ if-idf = 0.28( 0.03 * 9.21) 20
- 22. Dataset ﻪ Basic dataset ﻩ One million images most frequently clicked in a popular commercial image-search engine ﻩ (50K, 200K, 500K) ﻪ Ground truth ﻩ Manually labeled 780 partial-duplicate web image form 19 groups. ﻩ Evaluation dataset = basic dataset + ground truth ﻪ Query ﻩ 150 images from ground truth 22
- 23. Evaluation ﻪ Baseline ﻩ Bag-of-features approach with soft assignment[13] [13] J. Philbin, O. Chum, M. Isard, J. Sivic, and A. Zisserman. Lost in quantization: Improving particular object retrieval in large scale image databases. In CVPR, 2008. 23
- 24. ﻪ Compare(HE) ﻩ enhance the with hamming embedding [3] by adding a 24-bit hamming code to filter out target features. [3] H. Jegou, M. Douze, and C. Schmid. Hamming embedding and weak geometric consistency for large scale image search. In ECCV, 2008. 24
- 25. baseline0.35 to Bundled(mem)0.40 a 14% improvement baseline0.35 to Bundled 0.49 a 40% improvement baseline0.35 to Bundled+HE0.52 a 49% improvement 25
- 26. ﻪ Compare(Re-ranking) ﻩ Full geometric verification, RANSAC for top 300 candidate images 26
- 27. Baseline+re-rank 0.50 to Bundled+re-rank 0.62 a 24% improvement Baseline 0.35 to Bundled+re-rank 0.62 a 77% improvement 27
- 28. ﻪ Trade-off ﻪ Run time ﻩ a single CPU on a 3.0GHz Core Duo desktop with 16G memory 28
- 29. Sample results Query Image Baseline approach Our approach 29
- 30. 30
- 31. CONCLUSION 31
- 32. Conclusion ﻪ Bundled features property ﻩ More discriminative than individual SIFT features. ﻩ Simple and robust geometric constraints ﻩ Partially match two groups of SIFT features ﻪ Advantage ﻩ Robustness to occlusion, photometric and geometric changes 32
- 33. END Thanks for your Listening