Results 1  10
of
13
Heterogeneous embedding for subjective artist similarity
 In Tenth International Symposium for Music Information Retrieval (ISMIR2009
, 2009
"... We describe an artist recommendation system which integrates several heterogeneous data sources to form a holistic similarity space. Using social, semantic, and acoustic features, we learn a lowdimensional feature transformation which is optimized to reproduce humanderived measurements of subjecti ..."
Abstract

Cited by 17 (1 self)
 Add to MetaCart
We describe an artist recommendation system which integrates several heterogeneous data sources to form a holistic similarity space. Using social, semantic, and acoustic features, we learn a lowdimensional feature transformation which is optimized to reproduce humanderived measurements of subjective similarity between artists. By producing lowdimensional representations of artists, our system is suitable for visualization and recommendation tasks. 1.
Partial Order Embedding with Multiple Kernels
"... We consider the problem of embedding arbitrary objects (e.g., images, audio, documents) into Euclidean space subject to a partial order over pairwise distances. Partial order constraints arise naturally when modeling human perception of similarity. Our partial order framework enables the use of grap ..."
Abstract

Cited by 9 (4 self)
 Add to MetaCart
We consider the problem of embedding arbitrary objects (e.g., images, audio, documents) into Euclidean space subject to a partial order over pairwise distances. Partial order constraints arise naturally when modeling human perception of similarity. Our partial order framework enables the use of graphtheoretic tools to more efficiently produce the embedding, and exploit global structure within the constraint set. We present an embedding algorithm based on semidefinite programming, which can be parameterized by multiple kernels to yield a unified space from heterogeneous features. 1.
Adaptively Learning the Crowd Kernel
"... We introduce an algorithm that, given n objects, learns a similarity matrix over all n2 pairs, from crowdsourced data alone. The algorithm samples responses to adaptively chosen tripletbased relativesimilarity queries. Each query has the form “is object a more similar to b or to c? ” and is chosen ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
We introduce an algorithm that, given n objects, learns a similarity matrix over all n2 pairs, from crowdsourced data alone. The algorithm samples responses to adaptively chosen tripletbased relativesimilarity queries. Each query has the form “is object a more similar to b or to c? ” and is chosen to be maximally informative given the preceding responses. The output is an embedding of the objects into Euclidean space (like MDS); we refer to this as the “crowd kernel. ” SVMs reveal that the crowd kernel captures prominent and subtle features across a number of domains, such as “is striped ” among neckties and “vowel vs. consonant ” among letters. 1.
Learning Multimodal Similarity
"... In many applications involving multimedia data, the definition of similarity between items is integral to several key tasks, including nearestneighbor retrieval, classification, and recommendation. Data in such regimes typically exhibits multiple modalities, such as acoustic and visual content of ..."
Abstract

Cited by 8 (1 self)
 Add to MetaCart
In many applications involving multimedia data, the definition of similarity between items is integral to several key tasks, including nearestneighbor retrieval, classification, and recommendation. Data in such regimes typically exhibits multiple modalities, such as acoustic and visual content of video. Integrating such heterogeneous data to form a holistic similarity space is therefore a key challenge to be overcome in many realworld applications. We present a novel multiple kernel learning technique for integrating heterogeneous data into a single, unified similarity space. Our algorithm learns an optimal ensemble of kernel transformations which conform to measurements of human perceptual similarity, as expressed by relative comparisons. To cope with the ubiquitous problems of subjectivity and inconsistency in multimedia similarity, we develop graphbased techniques to filter similarity measurements, resulting in a simplified and robust training procedure.
Linear Embeddings in NonRigid Structure from Motion
"... This paper proposes a method to recover the embedding of the possible shapes assumed by a deforming nonrigid object by comparing triplets of frames from an orthographic video sequence. We assume that we are given features tracked with no occlusions and no outliers but possible noise, an orthographic ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
This paper proposes a method to recover the embedding of the possible shapes assumed by a deforming nonrigid object by comparing triplets of frames from an orthographic video sequence. We assume that we are given features tracked with no occlusions and no outliers but possible noise, an orthographic camera and that any 3D shape of a deforming object is a linear combination of several canonical shapes. By exploiting any repetition in the object motion and defining an ordering between triplets of frames in a Generalized NonMetric MultiDimensional Scaling framework, our approach recovers the shape coefficients of the linear combination, independently from other structure and motion parameters. From this point, a good estimate of the remaining unknowns is obtained for a final optimization to perform full nonrigid structure from motion. Results are presented on synthetic and real image sequences and our method is found to perform better than current state of the art. 1.
Gsml: A unified framework for sparse metric learning
 In Data Mining, 2009. ICDM’09. Ninth IEEE International Conference on
, 2009
"... There has been significant recent interest in sparse metric learning (SML) in which we simultaneously learn both a good distance metric and a lowdimensional representation. Unfortunately, the performance of existing sparse metric learning approaches is usually limited because the authors assumed ce ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
There has been significant recent interest in sparse metric learning (SML) in which we simultaneously learn both a good distance metric and a lowdimensional representation. Unfortunately, the performance of existing sparse metric learning approaches is usually limited because the authors assumed certain problem relaxations or they target the SML objective indirectly. In this paper, we propose a Generalized Sparse Metric Learning method (GSML). This novel framework offers a unified view for understanding many of the popular sparse metric learning algorithms including the Sparse Metric Learning framework proposed in [15], the Large Margin Nearest Neighbor (LMNN) [21][22], and the Dranking Vector Machine (Dranking VM) [14]. Moreover, GSML also establishes a close relationship with the
Learning Dissimilarities by Ranking: From SDP to QP
"... We consider the problem of learning dissimilarities between points via formulations which preserve a specified ordering between points rather than the numerical values of the dissimilarities. Dissimilarity ranking (dranking) learns from instances like “A is more similar to B than C is to D ” or “The ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
We consider the problem of learning dissimilarities between points via formulations which preserve a specified ordering between points rather than the numerical values of the dissimilarities. Dissimilarity ranking (dranking) learns from instances like “A is more similar to B than C is to D ” or “The distance between E and F is larger than that between G and H”. Three formulations of dranking problems are presented and new algorithms are presented for two of them, one by semidefinite programming (SDP) and one by quadratic programming (QP). Among the novel capabilities of these approaches are outofsample prediction and scalability to large problems. 1.
Distance Metric Learning from Pairwise Proximities
"... We compare techniques for embedding a data set into Euclidean space under different notions of proximity constraints. 1 ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
We compare techniques for embedding a data set into Euclidean space under different notions of proximity constraints. 1
Perceptionbased Visual Quality Measures Georgia Albuquerque ∗
"... In recent years diverse quality measures to support the exploration of highdimensional data sets have been proposed. Such measures can be very useful to rank and select informationbearing projections of very high dimensional data, when the visual exploration of all possible projections becomes unf ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
In recent years diverse quality measures to support the exploration of highdimensional data sets have been proposed. Such measures can be very useful to rank and select informationbearing projections of very high dimensional data, when the visual exploration of all possible projections becomes unfeasible. But even though a ranking of the low dimensional projections may support the user in the visual exploration task, different measures deliver different distances between the views that do not necessarily match the expectations of human perception. As an alternative solution, we propose a perceptionbased approach that, similar to the existing measures, can be used to select information bearing projections of the data. Specifically, we construct a perceptual embedding for the different projections based on the data from a psychophysics study and multidimensional scaling. This embedding together with a ranking function is then used to estimate the value of the projections for a specific user task in a perceptual sense.
Sony Pictures Imageworks
"... We design and implement a comprehensive study of the perception of gloss. This is the largest study of its kind to date, and the first to use real material measurements. In addition, we develop a novel multidimensional scaling (MDS) algorithm for analyzing pairwise comparisons. The data from the ps ..."
Abstract
 Add to MetaCart
We design and implement a comprehensive study of the perception of gloss. This is the largest study of its kind to date, and the first to use real material measurements. In addition, we develop a novel multidimensional scaling (MDS) algorithm for analyzing pairwise comparisons. The data from the psychophysics study and the MDS algorithm is used to construct a low dimensional perceptual embedding of these bidirectional reflectance distribution functions (BRDFs). The embedding is validated by correlating it with nine gloss dimensions, fitted parameters of seven analytical BRDF models, and a perceptual parameterization of Ward’s model. We also introduce a novel perceptual interpolation scheme that uses the embedding to provide the user with an intuitive interface for navigating the space of gloss and constructing new materials.