E
Emanuele Coviello
Researcher at University of California, San Diego
Publications - 24
Citations - 2066
Emanuele Coviello is an academic researcher from University of California, San Diego. The author has contributed to research in topics: Generative model & Hidden Markov model. The author has an hindex of 13, co-authored 23 publications receiving 1803 citations. Previous affiliations of Emanuele Coviello include University of California & Amazon.com.
Papers
More filters
Proceedings ArticleDOI
A new approach to cross-modal multimedia retrieval
Nikhil Rasiwasia,Jose Costa Pereira,Emanuele Coviello,Gabriel Doyle,Gert R. G. Lanckriet,Roger Levy,Nuno Vasconcelos +6 more
TL;DR: It is shown that accounting for cross-modal correlations and semantic abstraction both improve retrieval accuracy and are shown to outperform state-of-the-art image retrieval systems on a unimodal retrieval task.
Journal ArticleDOI
On the Role of Correlation and Abstraction in Cross-Modal Multimedia Retrieval
Jose Costa Pereira,Emanuele Coviello,Gabriel Doyle,Nikhil Rasiwasia,Gert R. G. Lanckriet,Roger Levy,Nuno Vasconcelos +6 more
TL;DR: A mathematical formulation equating the design of cross-modal retrieval systems to that of isomorphic feature spaces for different content modalities is proposed, finding that both hypotheses hold, in a complementary form, although evidence in favor of the abstraction hypothesis is stronger than that for correlation.
Journal ArticleDOI
Clustering Dynamic Textures with the Hierarchical EM Algorithm for Modeling Video
TL;DR: This paper derives a new algorithm for clustering DT models that is based on the hierarchical EM algorithm, and demonstrates the efficacy of the clustering algorithm on several applications in motion analysis, including hierarchical motion clustering, semantic motion annotation, and bag-of-systems codebook generation.
Journal ArticleDOI
Time Series Models for Semantic Music Annotation
TL;DR: A novel approach to automatic music annotation and retrieval that captures temporal aspects as well as timbral content, and a novel, efficient, and hierarchical expectation-maximization algorithm for DTM (HEM-DTM) is used to summarize the common information shared by DTMs modeling individual songs associated with a tag.
Journal ArticleDOI
Clustering hidden Markov models with variational HEM
TL;DR: A novel algorithm to cluster HMMs based on the hierarchical EM (HEM) algorithm, which effectively leverages large amounts of data when learning annotation models by using an efficient hierarchical estimation procedure, which reduces learning times and memory requirements, while improving model robustness through better regularization.