Stand-Alone Objective Segmentation Quality Evaluation
- PDF / 1,053,667 Bytes
- 12 Pages / 600 x 792 pts Page_size
- 27 Downloads / 172 Views
Stand-Alone Objective Segmentation Quality Evaluation Paulo Lobato Correia Instituto Superior T´ecnico, Instituto de Telecomunic¸a˜oes, Av. Rovisco Pais, 1049-001 Lisboa, Portugal Email: [email protected]
Fernando Pereira Instituto Superior T´ecnico, Instituto de Telecomunic¸a˜oes, Av. Rovisco Pais, 1049-001 Lisboa, Portugal Email: [email protected] Received 31 July 2001 and in revised form 4 January 2002 The identification of objects in video sequences, that is, video segmentation, plays a major role in emerging interactive multimedia services, such as those enabled by the ISO MPEG-4 and MPEG-7 standards. In this context, assessing the adequacy of the identified objects to the application targets, that is, evaluating the segmentation quality, assumes a crucial importance. Video segmentation technology has received considerable attention in the literature, with algorithms being proposed to address various types of applications. However, the segmentation quality performance evaluation of those algorithms is often ad hoc, and a well-established solution is not available. In fact, the field of objective segmentation quality evaluation is still maturing; recently, some more efforts have been made, mainly following the emergence of the MPEG object-based coding and description standards. This paper discusses the problem of objective segmentation quality evaluation in its most difficult scenario: stand-alone evaluation, that is, when a reference segmentation is not available for comparative evaluation. In particular, objective metrics are proposed for the evaluation of stand-alone segmentation quality for both individual objects and overall segmentation partitions. Keywords and phrases: video segmentation, segmentation quality evaluation, objective segmentation quality, stand-alone segmentation quality evaluation.
1.
INTRODUCTION
With the publication of the MPEG-4 standard in the Spring of 1999 [1], which allows to independently encode audiovisual objects, and the development of the MPEG-7 standard [2], allowing the content-based description of audiovisual material, the MPEG committee has given a significant contribution for the development of a new generation of interactive multimedia services. Innovative types of interaction are often based on the understanding of a video scene as composed by a set of video objects, to which it is possible to associate specific information as well as interactive “hooks” to deploy the desired application behaviour. To enable such type of interactive services, an understanding of the scene semantics is required, notably in terms of the relevant objects that are present. It is in this context that video segmentation plays a determinant role. Segmentation may be automatically obtained at the video production stage, for example, by using chroma keying techniques, or it may have to be obtained from the images captured by a camera by using appropriate segmentation algorithms. The evaluation of the adequacy of a segmentation algorithm, and its parameters’ configuration, for a given
Data Loading...
