Seminar: "Exploiting large image sets for scene parsing"; September 4th, 2014 at 11:00
Speaker: José Álvarez, researcher at NICTA, Canberra, ACT, Australia
Abstract: There is increasing interest in exploiting multiple images for scene understanding, with great progress in areas such as cosegmentation and video segmentation. Jointly analyzing the images in a large set offers the opportunity to exploit a greater source of information than when considering a single image on its own. However, this also yields challenges, since, to effectively exploit all the available information, the resulting methods need to consider not just local connections, but efficiently analyze similarity between all pairs of pixels within and across all the images. In this paper, we propose to model an image set as a fully-connected pairwise Conditional Random Field (CRF) defined over the image pixels, or superpixels, with Gaussian edge potentials. We show that this lets us co-label the images of a large set efficiently, thus yielding increased accuracy at no additional computational cost compared to sequential labeling of the images. Furthermore, we show that our model can be applied to the semi-supervised case, where we jointly consider labeled and unlabeled data in the CRF. This allows us to either entirely bypass the time-consuming computation of unary terms, or to exploit unaries computed at sparse image locations. Our experimental evaluation demonstrates that our framework lets us handle over ten thousand images in a matter of seconds.
In the press: Xerox: “Wir bleiben den Druckern treu”
Published in Futurezone Technology News (27 Aug 2014)
CLEF 2014 Conference and Labs of the Evaluation Forum; 15 - 18 September 2014, Sheffield - UK
Abstract: In this paper we report the contribution of XRCE team to the Domain Adaptation Challenge organized in the framework of ImageCLEF 2014 competition. We describe our approach to build an image classification system when a weak image annotation in the target domain is compensated by massively annotated images in source domains. One method is based using several heterogeneous methods for the domain adaptation aimed at the late fusion of the individual predictions. One big class of domain adaptation methods addresses a selective reuse of instances from source domains for target domain. We adopt the adaptive boosting for weighting source instances which learns a combination of weak classifiers in the target domain. Another class of methods aims to transform both target and source domains in a common space. In this class we focused on metric learning approaches aimed at reducing distances between images from the same class and to increase distances of different classes independently if they are from source or target domain. Combined the above approaches with a ”brute-force” SVM-based approach we obtain a set of heterogeneous classifiers for class prediction of target instances. In order to improve the overall accuracy, we combine individual classifiers through different versions of majority voting. We describe different series of experiments including those submitted for the official competition and analyze their results.
ECCV 2014 – European Conference on Computer Vision; Zurich, September 6-12, 2014
Friday, September 12, 2014, full day workshop: Transferring and Adapting Source Knowledge (TASK) in Computer Vision
Workshop Organizers: Antonio M. Lopez (Computer Vision Center and Universitat Autònoma de Barcelona), Kate Saenko (University of Massachusetts Lowell), Francesco Orabona (Toyota Technological Institute Chicago), José Antonio Rodríguez (Xerox Research Europe), David Vázquez (Computer Vision Center), Sebastian Ramos (Computer Vision Center and Universitat Autònoma de Barcelona), Jiaolong Xu (Computer Vision Center and Universitat Autònoma de Barcelona)
Publication: Part of Speech Tagging for French Social Media Data
Authors: Farhad Nooralahzadeh, Caroline Brun, Claude Roux
Discover in this video what it’s like to work at Xerox Research Centre Europe.
Xerox Research Centre Europe (XRCE) is a young, dynamic research organization, which creates innovative technologies to support growth in Xerox business process outsourcing and document management services businesses.
Our domains of research stretch from the social sciences to computer science. We have renowned expertise in machine learning, natural language processing, computer vision, ethnography and services computing.
XRCE is part of the Xerox Innovation group made up of 650 researchers and engineers in four world-renowned research and technology centres. Our goal is to make Xerox a great place to work. Through a comprehensive set of employee-focused initiatives, we promote diversity by nurturing a culture of inclusion and opportunity, and through measurable actions.
Xerox complies with Equal Employment Opportunity (EEO) guidelines. Xerox does not discriminate on the basis of race, colour, religious belief, sex, age, national origin, citizenship status, marital status, union status, sexual orientation, gender identity or individuals with a disability. Our main building is compliant with the AGEFIPH standards for disabled access.
The Grenoble site is set in a park in the heart of the French Alps in a stunning location only a few kilometers from the city centre. The city of Grenoble has a large scientific community made up of national research institutes (CNRS, Universities, INRIA) and private industries. Stimulated also by the presence of a large student community, Grenoble has become a resolutely modern city, with a rich heritage and a vibrant cultural scene. It is a lively and cosmopolitan place, offering a host of leisure opportunities. Winter sports resorts just half an hour from campus and three natural parks at the city limits make running, skiing, trekking, climbing and paragliding easily available.
Grenoble is close to both the Swiss and Italian borders.
AEPI, the Grenoble-Isere Economic Development Agency: "Grenoble-Isere, the place to innovate, where the only limit is your imagination."
Xerox Search Opportunities