Graph-Cut Transducers for Relevance Feedback in Content Based Image Retrieval

Sahbi, Hichem; Audibert, Jean-Yves; Keriven, Renaud

doi:10.1109/iccv.2007.4408861

Cited by 24 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among the usual techniques, based on SVM or boosting, we preferred testing a transduction-based learning. Some author followed the same path (see [21][16] [15] [14]). The idea is to take advantage both of the unlabeled and labeled samples in a transductive inference manner, learning from an incremental amount of training samples (feedbacks, in this case).…”

Section: Related Workmentioning

confidence: 98%

See 1 more Smart Citation

Relevance feedback strategies for artistic image collections tagging

Grana

Borghesani

Cucchiara

2011

Proceedings of the 1st ACM International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

This paper provides an analysis on relevance feedback techniques in a multimedia system designed for the interactive exploration and annotation of artistic collections, in particular illuminated manuscripts. The relevance feedback is presented not only as a very effective technique to improve the performance of the system, but also as a clever way to increase the user experience, mixing the interactive surfing through the artistic content with the possibility to gather valuable information from the user, and consequently improving his retrieval satisfaction. We compare a modification of the Mean-Shift Feature Space Warping algorithm, as representative of the standard RF procedures, and a learning-based technique based on transduction, considered in order to overcome some limitation of the previous technique. Experiments are reported regarding the adopted visual features based on covariance matrices.

show abstract

Section: Related Workmentioning

confidence: 98%

“…Once converted in matrix notation, Eq. 14 becomes: (15) where L = D − W is the graph Laplacian. W is the weight matrix, while D is the matrix which represents the degree of vertices:…”

Section: Transductive Relevance Feed-backmentioning

confidence: 99%

Relevance feedback strategies for artistic image collections tagging

Grana

Borghesani

Cucchiara

2011

Proceedings of the 1st ACM International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

show abstract

“…Sahbi et al [14] employ a graph-cuts based approach for interactive retrieval. In a later work, Sahbi et.…”

Section: Related Workmentioning

confidence: 99%

Interactive retrieval of targets for wide area surveillance

Ali¹,

Javed²,

Haering

et al. 2010

Proceedings of the 18th ACM International Conference on Multimedia

View full text Add to dashboard Cite

We address the problem of interactive search for a target of interest in surveillance imagery. Our solution consists of iteratively learning a distance metric for retrieval, based on user feedback. The approach employs (retrieval) rank based constraints and convex optimization to efficiently learn the distance metric. The algorithm uses both user labeled and unlabeled examples in the learning process. The method is fast enough for a new metric to be learned interactively for each target query. In order to reduce the burden on the user, a model-independent active learning method is used to select key examples, for response solicitation. This leads to a significant reduction in the number of user-interactions required for retrieving the target of interest. The proposed method is evaluated on challenging pedestrian and vehicle data sets, and compares favorably to the state of the art in target re-acquisition algorithms.

show abstract

“…The goal is to improve the search results exploiting the notion of the user on the correctness of the automatically retrieved results. To this purpose we exploit transductive learning, a well-known approach in machine learning community [5,6,7], that recently gained attention in multimedia community too [8], with the final aim of incorporating such an inference element in the forensic query process. The paper proposes a system for interactive active querying on forensics data, in particular on people appearance and trajectories.…”

Section: Introductionmentioning

confidence: 99%

Iterative active querying for surveillance data retrieval in crime detection and forensics

Coppi

Calderara

Cucchiara

2011

4th International Conference on Imaging for Crime Detection and Prevention 2011 (ICDP 2011)

View full text Add to dashboard Cite

Large sets of visual data are now available both, in real time and off line, at time of investigation in multimedia forensics, however passive querying systems often encounter difficulties in retrieving significant results. In this paper we propose an iterative active querying system for video surveillance and forensic applications based on the continuous interaction between the user and the system. The positive and negative user feedbacks are exploited as the input of a graph based transductive procedure for iteratively refining the initial query results. Experiments are shown using people trajectories and people appearance as distance metrics.

show abstract

Graph-Cut Transducers for Relevance Feedback in Content Based Image Retrieval

Cited by 24 publications

References 14 publications

Relevance feedback strategies for artistic image collections tagging

Relevance feedback strategies for artistic image collections tagging

Interactive retrieval of targets for wide area surveillance

Iterative active querying for surveillance data retrieval in crime detection and forensics

Contact Info

Product

Resources

About