Mining Multiple Queries for Image Retrieval: On-the-Fly Learning of an Object-Specific Mid-level Representation

Fernando, Basura; Tuytelaars, Tinne

doi:10.1109/iccv.2013.316

Cited by 37 publications

(25 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Others [27,30] have re-weighted source samples based on target sample similarity. Saenko et al [35] map source and target data to a domain-invariant subspace; several later works build upon this idea [19,23,26,38]. For deep learning, Ganin and Lempitsky [20] incorporate a branch in their network architecture that classifies an input sample as being from one of two domains.…”

Section: Related Workmentioning

confidence: 99%

The Art of Detection

Crowley

Zisserman

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The objective of this work is to recognize object categories in paintings, such as cars, cows and cathedrals. We achieve this by training classifiers from natural images of the objects. We make the following contributions: (i) we measure the extent of the domain shift problem for image-level classifiers trained on natural images vs paintings, for a variety of CNN architectures; (ii) we demonstrate that classificationby-detection (i.e. learning classifiers for regions rather than the entire image) recognizes (and locates) a wide range of small objects in paintings that are not picked up by image-level classifiers, and combining these two methods improves performance; and (iii) we develop a system that learns a region-level classifier on-the-fly for an object category of a user's choosing, which is then applied to over 60 million object regions across 210,000 paintings to retrieve localised instances of that category.

show abstract

Section: Related Workmentioning

confidence: 99%

The Art of Detection

Crowley

Zisserman

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…In [22], Fernando and Tuytelaars focused on an unsupervised approach to discover visual patterns for the specific object, that are robust to noisy query images. By combining multiple queries in a principled manner, Arandjelović and Zisserman [23] described a framework where the query images are obtained by using textual Google image search.…”

Section: Introductionmentioning

confidence: 99%

“…In [15], Arandjelović and Zisserman proposed a discriminative query expansion approach(DQE), where an SVM classifier was trained by initial retrieved images and used to re-rank final images in dataset. In [27], Fernando and Tuytelaars designed a set of mid-level patterns based on the images retrieved by the Google image search engine. These patterns were used for improving the retrieval performance.…”

Section: Introductionmentioning

confidence: 99%

Finding more relevance: Propagating similarity on Markov random field for object retrieval

Peng

Zhu

et al. 2015

Signal Processing: Image Communication

View full text Add to dashboard Cite

“…FIM has begun to be applied to problems involving images. Here, visual word mining attempts to find meaningful co-occurring patterns by taking advantage of the power of FIM especially in the context of QE, and it has been used in video mining [10], visual phrase mining [11], mining of multiple queries [12], and mining for re-ranking and classification [13]. FIM is also reported to be able to find unambiguous spatially visual meanings [11].…”

Section: Related Workmentioning

confidence: 99%

Query Bootstrapping: A Visual Mining Based Query Expansion

Kasamwattanarote

Uchida

Satoh

2016

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYBag of Visual Words (BoVW) is an effective framework for image retrieval. Query expansion (QE) further boosts retrieval performance by refining a query with relevant visual words found from the geometric consistency check between the query image and highly ranked retrieved images obtained from the first round of retrieval. Since QE checks the pairwise consistency between query and highly ranked images, its performance may deteriorate when there are slight degradations in the query image. We propose Query Bootstrapping as a variant of QE to circumvent this problem by using the consistency of highly ranked images instead of pairwise consistency. In so doing, we regard frequently co-occurring visual words in highly ranked images as relevant visual words. Frequent itemset mining (FIM) is used to find such visual words efficiently. However, the FIM-based approach requires sensitive parameters to be fine-tuned, namely, support (min/max-support) and the number of top ranked images (top-k). Here, we propose an adaptive support algorithm that adaptively determines both the minimum support and maximum support by referring to the first round's retrieval list. Selecting relevant images by using a geometric consistency check further boosts retrieval performance by reducing outlier images from a mining process. An important parameter for the LO-RANSAC algorithm that is used for the geometric consistency check, namely, inlier threshold, is automatically determined by our algorithm. We further introduce tf-fi-idf on top of tf-idf in order to take into account the frequency of inliers (fi) in the retrieved images. We evaluated the performance of QB in terms of mean average precision (mAP) on three benchmark datasets and found that it gave significant performance boosts of 5.37%, 9.65%, and 8.52% over that of state-of-the-art QE on Oxford 5k, Oxford 105k, and Paris 6k, respectively. key words: image retrieval, instance search, query expansion, frequent itemset mining, visual word mining, query bootstrapping, adaptive support, adaptive inlier threshold

show abstract

Mining Multiple Queries for Image Retrieval: On-the-Fly Learning of an Object-Specific Mid-level Representation

Abstract: In this paper we present a new method for object retrieval starting from multiple query images. The

Cited by 37 publications

References 27 publications

The Art of Detection

The Art of Detection

Finding more relevance: Propagating similarity on Markov random field for object retrieval

Query Bootstrapping: A Visual Mining Based Query Expansion

Contact Info

Product

Resources

About