Ontology-based video retrieval

Lecture Notes in Computer Science, 2004

Domain ontologies are very useful for indexing, query specification, retrieval and filtering, user interfaces, even information extraction from audiovisual material. The dominant emerging language standard for the description of domain ontologies is OWL. We describe here a methodology and software that we have developed for the interoperability of OWL with the complete MPEG-7 MDS so that domain ontologies described in OWL can be transparently integrated with the MPEG-7 MDS metadata. This allows applications that recognize and use the MPEG-7 MDS constructs to make use of domain ontologies for applications like indexing, retrieval, filtering etc. resulting in more effective user retrieval and interaction with audiovisual material.

2006

Abstract Effective usage of multimedia digital libraries has to deal with the problem of building efficient content annotation and retrieval tools. MOM (Multimedia Ontology Manager) is a complete system that allows the creation of multimedia ontologies, supports automatic annotation and creation of extended text (and audio) commentaries of video sequences, and permits complex queries by reasoning on the ontology.

Nowadays, the video documents like educational courses available on the web increases significantly. However, the information retrieval systems today can not return to the users (students or teachers) of parts of those videos that meet their exact needs expressed by a query consisting of semantic information. In this paper, we present a model of pedagogical knowledge of current videos. This knowledge is used throughout the process of indexing and semantic search segments instructional videos. Our experimental results show that the proposed approach is promising.

Proceedings of the 13th annual ACM international conference on Multimedia - MULTIMEDIA '05, 2005

To ensure access to growing video collections, annotation is becoming more and more important. Using background knowledge in the form of ontologies or thesauri is a way to facilitate annotation in a broad domain. Current ontologies are not suitable for (semi-) automatic annotation of visual resources as they contain little visual information about the concepts they describe. We investigate how an ontology that does contain visual information can facilitate annotation in a broad domain and identify requirements that a visual ontology has to meet. Based on these requirements, we create a visual ontology out of two existing knowledge corpora (WordNet and MPEG-7) by creating links between visual and general concepts. We test performance of the ontology on 40 shots of news video, and discuss the added value of each visual property.

—As the multimedia content over the internet is increasing day by day, efficient methods for retrieval of these huge amount of data is required. Video annotation is one of the widely used methods to analyze and retrieve these huge video data. The process of video annotation is complicated as it requires a large amount of processing to analyze the contents in the video. This paper introduces a video annotation system based on ontology with HoG features given for training the classifiers. The SIFT and HoG features of the images are extracted and used for training the classifiers for doing a comparison on performance of the classifiers. An analysis of the results is done to find which feature is better to train the classifier for getting more prominent annotated video database. Retrieval of the videos based on objects from the annotated video database is also done.

Recent advances in digital video analysis and retrieval have made video more accessible than ever. The representation and recognition of events in a video is important for a number of tasks such as video surveillance, video browsing and content based video indexing. Raw data and low-level features alone are not sufficient to fulfill the user's needs; that is, a deeper understanding of the content at the semantic level is required .Currently, manual techniques, which are inefficient, subjective and costly in time and limit the querying capabilities .Here, we propose a semantic content extraction system that allows the user to query and retrieve objects, events, and concepts that are extracted automatically. We introduce an ontology-based fuzzy video semantic content model that uses spatial/temporal relations in event and concept definitions. This metaontology definition provides a wide-domain applicable rule construction standard that allows the user to construct ontology for a given domain. In addition to domain ontologies, we use additional rule definitions (without using ontology) to define some complex situations more effectively. The proposed framework has been fully implemented and tested on three different domains and it provides satisfactory results.

Procedia Computer Science, 2016

Numerous educational video lectures, CCTV surveillance, transport and other types have upgraded the impact of multimedia video content. In order to make large video databases realistic, video data has to be automatically indexed in order to search and retrieve relevant material. An annotation is a markup reference made to data in video in order to improve the video accessibility. Video annotation is used to examine the massive quantity of multimedia data in the repositories. Video annotation refers to the taking out of significant data present in video and placing this data to the video can benefit in "retrieval, browsing, analysis, searching comparison and categorization". Video annotation implies taking out of data and to attach such metadata to the video which will "accelerate the retrieval speed, ease of access, analysis and categorization". It permits fast and better understanding of video content and improves the performance of retrieval and decreases human time & efforts for better study of videos. Video annotation is imperative technique that assists in video access. Proposed system provides effortless access to the data of the video and decrease the time necessary to access and evaluate the video. Ontology-based video annotation helps the user to get the semantic information from video, which is essential to search the needful data from a video.

Multimedia Tools and Applications, 2008

In this paper we present a framework for unified, personalized access to heterogeneous multimedia content in distributed repositories. Focusing on semantic analysis of multimedia documents, metadata, user queries and user profiles, it contributes to the bridging of the gap between the semantic nature of user queries and raw multimedia documents. The proposed approach utilizes as input visual content analysis results, as well as analyzes and exploits associated textual annotation, in order to extract the underlying semantics, construct a semantic index and classify documents to topics, based on a unified knowledge and semantics representation model. It may then accept user queries, and, carrying out semantic interpretation and expansion, retrieve documents from the index and rank them according to user preferences, similarly to text retrieval. All processes are based on a novel semantic processing methodology, employing fuzzy algebra and principles of taxonomic knowledge representation. The first part of this work presented in this paper deals with data and knowledge models, manipulation of multimedia content annotations and semantic indexing, while the second part will continue on the use of the extracted semantic information for personalized retrieval.

… information retrieval on …, 2007

This paper proposes a new method for using implicit user feedback from clickthrough data to provide personalized ranking of results in a video retrieval system. The annotation based search is complemented with a feature based ranking in our approach. The ranking algorithm uses belief revision in a Bayesian Network, which is derived from a multimedia ontology that captures the probabilistic association of a concept with expected video features. We have developed a content model for videos using discrete feature states to enable Bayesian reasoning and to alleviate on-line feature processing overheads. We propose a reinforcement learning algorithm for the parameters of the Bayesian Network with the implicit feedback obtained from the clickthrough data.

2010

Abstract In this technical demonstration we present a novel web-based tool that allows a user friendly semantic browsing of video collections, based on ontologies, concepts, concept relations and concept clouds. The system is developed as a Rich Internet Application (RIA) to achieve a fast responsiveness and ease of use that can not be obtained by other web application paradigms, and uses streaming to access and inspect the videos.