Face attribute prediction has important applications in video surveillance, face retrieval, and social media. While a number of methods have been proposed for face attribute prediction, most of them did not explicitly...
详细信息
Face attribute prediction has important applications in video surveillance, face retrieval, and social media. While a number of methods have been proposed for face attribute prediction, most of them did not explicitly consider the attribute correlation and heterogeneity during feature learning. In this paper, we propose a Deep Multi-Task Learning (DMTL) network to jointly learn multiple models; each addresses the prediction of one category of homogenous attributes. Specifically, we group the heterogeneous face attributes into two categories (i.e., nominal and ordinal), and design corresponding prediction models. At the same time, we use a convolutional neural network (CNN) for early stage feature learning, which is shared by all the attributes. Experiments on the public-domain MORPH ii, CelebA, and LFWA databases show that the proposed approach outperforms the state of the art in joint face attribute prediction, and has good generalization ability.
databases are increasingly being used to store multi-media objects such as maps, images, audio and video. storage and retrieval of these objects is accomplished using multi-dimensional index structures such as R*-tree...
ISBN:
(纸本)9780897919951
databases are increasingly being used to store multi-media objects such as maps, images, audio and video. storage and retrieval of these objects is accomplished using multi-dimensional index structures such as R*-trees and SS-trees. As dimensionality increases, query performance in these index structures degrades. This phenomenon, generally referred to as the dimensionality curse, can be circumvented by reducing the dimensionality of the data. Such a reduction is however accompanied by a loss of precision of query results. Current techniques such as QBIC use SVD transform-based dimensionality reduction to ensure high query precision. The drawback of this approach is that SVD is expensive to compute, and therefore not readily applicable to dynamic databases. In this paper, we propose novel techniques for performing SVD-based dimensionality reduction in dynamic databases. When the data distribution changes considerably so as to degrade query precision, we recompute the SVD transform and incorporate it in the existing index structure. For recomputing the SVD-transform, we propose a novel technique that uses aggregate data from the existing index rather than the entire data. This technique reduces the SVD-computation time without compromising query precision. We then explore efficient ways to incorporate the recomputed SVD-transform in the existing index structure without degrading subsequent query response times. These techniques reduce the computation time by a factor of 20 in experiments on color and texture image vectors. The error due to approximate computation of SVD is less than 10%.
We introduce a tuple-based model for specification of events among media segments in a multimedia presentation. A multimedia presentation contains various kinds of multimedia objects such as audio, video, images, and ...
详细信息
We introduce a tuple-based model for specification of events among media segments in a multimedia presentation. A multimedia presentation contains various kinds of multimedia objects such as audio, video, images, and so on. The organization of a presentation is a complex task in that the display order of presentation content (in time and space) must be specified. The critical decisions for presentation construction include content selection, content organization, and content delivery. Once the decision is made on the organization of the presentation content, it must be conveyed to the end user in the correct organizational order and in a timely fashion. Assuming that a set of multimedia segments gets organized into a presentation graph, this conceptual model formalizes a complete presentation using event-points for these segments. End-user environments of multimedia presentation are considered for the playout such as (i) the one without any constraint, (ii) another one with a single constraint, and (iii) the third one with multiple constraints (one constraint for each type of segments in the organized presentation). In accordance with these limitations (without violating any of the specified constraints), three methods are given for constructed presentations to be presented.
暂无评论