Scalable Recognition with a Vocabulary Tree

Nistér, D.; Stewénius, H.

doi:10.1109/cvpr.2006.264

Cited by 2,890 publications

(2,749 citation statements)

References 16 publications

Supporting

Mentioning

2,726

Contrasting

Unclassified

Order By: Relevance

“…We measure the retrieval error rate as the percentage of query images not correctly retrieved and verified with our pipeline. We briefly describe our retrieval pipeline, which is similar to other the state-of-the-art systems, such as [23,24,25,26].…”

Section: Database Retrievalmentioning

confidence: 99%

“…Using these descriptors, we train a 10 6 leaf, 6 level, vocabulary tree [23]. We use symmetric KL-divergence as the distance measure for both training and querying, since it performs better than L 2 -norm for HoG descriptors [27].…”

Section: Database Retrievalmentioning

confidence: 99%

“…We also use the weighting scheme proposed by Nistér [23], which reduces the contribution of less descriminative descriptors. Once the best 50 images are selected from the TF-IDF voting we perform pairwise matching with a ratio-test and a geometric consistency check.…”

Section: Database Retrievalmentioning

confidence: 99%

See 2 more Smart Citations

Unified Real-Time Tracking and Recognition with Rotation-Invariant Fast Features

Takacs

Chandrasekhar

Tsai

et al. 2010

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

111

View full text Add to dashboard Cite

We present a method that unifies tracking and video content recognition with applications to Mobile Augmented Reality (MAR). We introduce the Radial Gradient Transform (RGT) and an approximate RGT, yielding the Rotation-Invariant, Fast Feature (RIFF) descriptor. We demonstrate that RIFF is fast enough for real-time tracking, while robust enough for large scale retrieval tasks. At 26× the speed, our trackingscheme obtains a more accurate global affine motionmodel than the Kanade Lucas Tomasi (KLT) tracker. The same descriptors can achieve 94% retrieval accuracy from a database of 10 4 images.

show abstract

Section: Database Retrievalmentioning

confidence: 99%

Section: Database Retrievalmentioning

confidence: 99%

See 1 more Smart Citation

Unified Real-Time Tracking and Recognition with Rotation-Invariant Fast Features

Takacs

Chandrasekhar

Tsai

et al. 2010

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

111

View full text Add to dashboard Cite

show abstract

“…To build a vocabulary of visual words, interest regions in the images are detected with Hessian-affine detector [8], which provides good performance [9] and is widely used in visual word-based studies because of its insensitiveness to affine transformations such as scaling, reflection, rotation, etc. These regions are described in 128-dimension SIFT descriptors and then clustered by a hierarchical k-means algorithm [10], each cluster representing a visual word. Then each image is represented in a bag of visual words.…”

Section: B Detecting Reused Visual Elementsmentioning

confidence: 99%

Managing Presentation Slides with Reused Elements

Zhang¹,

Xiao²,

Hu³

et al. 2016

IJIET

View full text Add to dashboard Cite

Abstract-Slide presentations have become a ubiquitous tool for business and educational purposes. Instead of starting from scratch, slide composers tend to make new presentation slides by reusing materials from existing slides. Understanding how slide elements are copied from one presentation file to another and how presentation files are related to each other are difficult tasks. In this paper, it is investigated the management of multiple presentation files based on reused slide elements. Techniques are developed to detect textual and visual elements that have been reused across multiple presentation files. Interactive visualization methods are proposed to facilitate understanding the process by which these elements are reused and the relationship between the files that use them. A system with a user-friendly interface is designed, based on which experiments are performed to evaluate the effectiveness of the proposed methods.Index Terms-Presentation slide management, slide element reuse, slide element visualization.

show abstract

“…These features are robust to arbitrary changes in viewpoints. Then, hierarchical k-means clustering [19] is applied to the features, to group them based on their similarity. The result of the hierarchical clustering is used for the fast approximation of the nearest neighbor search, to efficiently resolve feature matching.…”

Section: Object Duplicate Detectionmentioning

confidence: 99%

Geotag propagation in social networks based on user trust model

Ivanov

Vajda

Lee

et al. 2010

Multimed Tools Appl

View full text Add to dashboard Cite

In the past few years sharing photos within social networks has become very popular. In order to make these huge collections easier to explore, images are usually tagged with representative keywords such as persons, events, objects, and locations. In order to speed up the time consuming tag annotation process, tags can be propagated based on the similarity between image content and context. In this paper, we present a system for efficient geotag propagation based on a combination of object duplicate detection and user trust modeling. The geotags are propagated by training a graph based object model for each of the landmarks on a small tagged image set and finding its duplicates within a large untagged image set. Based on the established correspondences between these two image sets and the reliability of the user, tags are propagated from the tagged to the untagged images. The user trust modeling reduces the risk of propagating wrong tags caused by spamming or faulty annotation. The effectiveness of the proposed method is demonstrated through a set of experiments on an image database containing various landmarks.

show abstract

Scalable Recognition with a Vocabulary Tree

Cited by 2,890 publications

References 16 publications

Unified Real-Time Tracking and Recognition with Rotation-Invariant Fast Features

Unified Real-Time Tracking and Recognition with Rotation-Invariant Fast Features

Managing Presentation Slides with Reused Elements

Geotag propagation in social networks based on user trust model

Contact Info

Product

Resources

About