Institution
Yahoo!
Company•London, United Kingdom•
About: Yahoo! is a company organization based out in London, United Kingdom. It is known for research contribution in the topics: Population & Web search query. The organization has 26749 authors who have published 29915 publications receiving 732583 citations. The organization is also known as: Yahoo! Inc. & Maudwen-Yahoo! Inc.
Papers published on a yearly basis
Papers
More filters
••
23 Jun 2014TL;DR: An extensive evaluation of the method compared to previous state-of-the-art approaches on the challenging PASCAL VOC 2007 and Object Discovery datasets and a large-scale study of co-localization on ImageNet, involving ground-truth annotations for 3, 624 classes and approximately 1 million images.
Abstract: In this paper, we tackle the problem of co-localization in real-world images. Co-localization is the problem of simultaneously localizing (with bounding boxes) objects of the same class across a set of distinct images. Although similar problems such as co-segmentation and weakly supervised localization have been previously studied, we focus on being able to perform co-localization in real-world settings, which are typically characterized by large amounts of intra-class variation, inter-class diversity, and annotation noise. To address these issues, we present a joint image-box formulation for solving the co-localization problem, and show how it can be relaxed to a convex quadratic program which can be efficiently solved. We perform an extensive evaluation of our method compared to previous state-of-the-art approaches on the challenging PASCAL VOC 2007 and Object Discovery datasets. In addition, we also present a large-scale study of co-localization on ImageNet, involving ground-truth annotations for 3, 624 classes and approximately 1 million images.
206 citations
••
11 Oct 2006TL;DR: This work presents a framework for the identification of user’s interest in an automatic way, based on the analysis of query logs, and establishes that the combination of supervised and unsupervised learning is a good alternative to find user‘s goals.
Abstract: The identification of the user’s intention or interest through queries that they submit to a search engine can be very useful to offer them more adequate results. In this work we present a framework for the identification of user’s interest in an automatic way, based on the analysis of query logs. This identification is made from two perspectives, the objectives or goals of a user and the categories in which these aims are situated. A manual classification of the queries was made in order to have a reference point and then we applied supervised and unsupervised learning techniques. The results obtained show that for a considerable amount of cases supervised learning is a good option, however through unsupervised learning we found relationships between users and behaviors that are not easy to detect just taking the query words. Also, through unsupervised learning we established that there are categories that we are not able to determine in contrast with other classes that were not considered but naturally appear after the clustering process. This allowed us to establish that the combination of supervised and unsupervised learning is a good alternative to find user’s goals. From supervised learning we can identify the user interest given certain established goals and categories; on the other hand, with unsupervised learning we can validate the goals and categories used, refine them and select the most appropriate to the user’s needs.
206 citations
••
TL;DR: The isolation of a previously unknown reovirus from a 39-year-old male patient in Melaka, Malaysia, who was suffering from high fever and acute respiratory disease at the time of virus isolation indicates that bat-borne reoviruses can be transmitted to and cause clinical diseases in humans.
Abstract: Respiratory infections constitute the most widespread human infectious disease, and a substantial proportion of them are caused by unknown etiological agents. Reoviruses (respiratory enteric orphan viruses) were first isolated from humans in the early 1950s and so named because they were not associated with any known disease. Here, we report a previously unknown reovirus (named "Melaka virus") isolated from a 39-year-old male patient in Melaka, Malaysia, who was suffering from high fever and acute respiratory disease at the time of virus isolation. Two of his family members developed similar symptoms approximately 1 week later and had serological evidence of infection with the same virus. Epidemiological tracing revealed that the family was exposed to a bat in the house approximately 1 week before the onset of the father's clinical symptoms. Genome sequence analysis indicated a close genetic relationship between Melaka virus and Pulau virus, a reovirus isolated in 1999 from fruit bats in Tioman Island, Malaysia. Screening of sera collected from human volunteers on the island revealed that 14 of 109 (13%) were positive for both Pulau and Melaka viruses. This is the first report of an orthoreovirus in association with acute human respiratory diseases. Melaka virus is serologically not related to the different types of mammalian reoviruses that were known to infect humans asymptomatically. These data indicate that bat-borne reoviruses can be transmitted to and cause clinical diseases in humans.
206 citations
••
TL;DR: There was a good interobserver reliability for the evaluation of patellar height according to the common radiological ratios, however, the high frequency of differing results between the different radiographic ratios showed that patella height classification as “alta,”“norma” or “baja” depends heavily on the chosen index.
Abstract: This study evaluated the reliability and interobserver variability of five patellar height ratios as measured by two examiners on standard radiographs: Insall-Salvati (IS), modified Insall-Salvati (MIS), Blackburne-Peel (BP), Caton-Deschamps (CD), and Labelle-Laurin (LL). Plain lateral radiographs with a knee flexion angle of 20° for IS, MIS, BP, and CD ratios and 90° for the LL method of 22 knees of 21 patients with varying pathological knee conditions were analyzed. Statistical results revealed a low interobserver variability with high correlation coefficients (0.86 for IS, 0.82 for MIS, 0.86 for BP, 0.92 for CD, and 0.81 for LL; P > 0.3) and low mean interobserver errors. However, regarding the reliability of the radiographic results of the different methods for patella alta, baja, or norma we found varying results in 68% of the patients. In two patients the patellar height was classified as alta, norma, or baja depending on the ratio used. Regarding the definitions of patellar height used by the authors of these methods, we found the lowest number of normal patellae with the IS ratio and no patella alta for the CD ratio. The LL method revealed the highest number of patella alta. The BP ratio showed intermediate results for both patella alta and baja, being the most moderate method. This study showed that there was a good interobserver reliability for the evaluation of patellar height according to the common radiological ratios. However, the high frequency of differing results between the different radiographic ratios showed that patellar height classification as “alta,”“norma,” or “baja” depends heavily on the chosen index. The differing results were due mainly to the normative patellar height data and to anatomical differences. Based on these findings we recommend a ratio using the articular surface of the patella in relation to the joint line. We recommend the BP method because it revealed the lowest interobserver variability and discriminated best among the groups alta, norma, and baja.
205 citations
••
16 Apr 2013TL;DR: This work presents a scalable approach to automatically suggest relevant clothing products, given a single image without metadata, and achieves clothing detection performance comparable to the state-of-the-art on a very recent annotated dataset, while being more than 50 times faster.
Abstract: We present a scalable approach to automatically suggest relevant clothing products, given a single image without metadata. We formulate the problem as cross-scenario retrieval: the query is a real-world image, while the products from online shopping catalogs are usually presented in a clean environment. We divide our approach into two main stages: a) Starting from articulated pose estimation, we segment the person area and cluster promising image regions in order to detect the clothing classes present in the query image. b) We use image retrieval techniques to retrieve visually similar products from each of the detected classes. We achieve clothing detection performance comparable to the state-of-the-art on a very recent annotated dataset, while being more than 50 times faster. Finally, we present a large scale clothing suggestion scenario, where the product database contains over one million products.
204 citations
Authors
Showing all 26766 results
Name | H-index | Papers | Citations |
---|---|---|---|
Ashok Kumar | 151 | 5654 | 164086 |
Alexander J. Smola | 122 | 434 | 110222 |
Howard I. Maibach | 116 | 1821 | 60765 |
Sanjay Jain | 103 | 881 | 46880 |
Amirhossein Sahebkar | 100 | 1307 | 46132 |
Marc Davis | 99 | 412 | 50243 |
Wenjun Zhang | 96 | 976 | 38530 |
Jian Xu | 94 | 1366 | 52057 |
Fortunato Ciardiello | 94 | 695 | 47352 |
Tong Zhang | 93 | 414 | 36519 |
Michael E. J. Lean | 92 | 411 | 30939 |
Ashish K. Jha | 87 | 503 | 30020 |
Xin Zhang | 87 | 1714 | 40102 |
Theunis Piersma | 86 | 632 | 34201 |
George Varghese | 84 | 253 | 28598 |