Feature based retrieval for animation video
03 Aug 2012-pp 905-909
TL;DR: The use of color feature information when used for the animation video search system and how the similarity between the query video clip and each video clip in database is computed.
Abstract: Numerous researches have been conducted for content based representation, analysis and retrieval but mainly for the professionally edited videos such as news, & sports videos not on animation video. The objective is to adapt existing image based information retrieval technique to work efficiently on animation video. This paper shows the use of color feature information when used for the animation video search system. In this proposed work, video clip is accepted as query-by-example and matched video clips are retrieved from animation video database. During the retrieval the color feature is extracted for input video clip using RGB2GRAY, RGB2HSV and RGB2YCBCR color space algorithm and compared with feature vector database in order to rank each video clip according to its distance to the query. We use the Euclidean Distance, City Block Distance and Canberra Distance measures to compute the similarity between the query video clip and each video clip in database. We report experimental result and validate the method on new large database of real animation video.
Citations
More filters
01 Dec 2019
TL;DR: This paper presents a new challenge on proposing a video recommendation system based on content using objects and features with the ability to search or block specific scenes.
Abstract: When it comes to searching online, massive information is available, it is really hard to provide relevant information to users based on their interest. Although while searching for data based on user inputs, they need to search the entire database, which is also very frustrating and time-consuming. Video consumption becoming essential in most users' life. On the most video platforms, users get their recommended videos based on some algorithms, calculations, implicit feed-backs, watch, search behaviors and search history. New videos suffer from cold-start which happens to freshly uploaded videos in which no data or reviews are available. Therefore, it is not easy to recommend these videos to some users. Another real problem that users face every day is that finding the desired content depends on the video being labeled or has multiple views. The search engine will find the videos based on keywords or tags, not on the content inside the video. One of the solutions for this problem is recommending videos based on content. This paper presents a new challenge on proposing a video recommendation system based on content using objects and features with the ability to search or block specific scenes.
3Â citations
Cites methods from "Feature based retrieval for animati..."
...An objection to accustom an image established information retrieval method that already exists is presented in [26]....
[...]
TL;DR: A recommendation system by content is proposed, the system detects the objects and sounds inside the video, and also adds the feature to search using uploaded scenes or filter scenes based on keyword inputted.
Abstract: Article history: Received 20 Jan 2022 Revised 08 Feb 2022 Accepted 08 Feb 2022 Available online In a world full of online videos, it is really hard to find relevant content as the data is simply too much. A recommendation system was created to refine this experience, to match relevant content to an interested user. Most recommending systems use algorithms, calculations, and implicit feedback. These methods are effective unless the video does not have implicit feedback in which the algorithms will mostly fail to get relevant content. This is known as cold-start that affects newly uploaded videos, since they start without any data or user comments. Another problem facing users every day is finding the content they want, because it is dependent on videos having labels or having many user views. Since the search engine's mechanism uses the tags and keywords inserted for the video rather than the actual content in it. In this paper, a recommendation system by content is proposed, the system detects the objects and sounds inside the video, and also adds the feature to search using uploaded scenes or filter scenes based on keyword inputted. More experimental results have been done with various scenarios to demonstrate the effectiveness of the proposed system in terms of video recommendation by content.
2Â citations
References
More filters
TL;DR: The major themes covered by the study include shot segmentation, key frame extraction, feature extraction, clustering, indexing and video retrieval-by similarity, probabilistic, transformational, refinement and relevance feedback.
Abstract: This study surveys current trends/methods in video retrieval. The major themes covered by the study include shot segmentation, key frame extraction, feature extraction, clustering, indexing and video retrieval-by similarity, probabilistic, transformational, refinement and relevance feedback. This work has done in an aim to assist the upcoming researchers in the field of video retrieval, to know about the techniques and methods available for video retrieval.
117Â citations
"Feature based retrieval for animati..." refers background in this paper
...Both the algorithms represent an image as a vector of features and compute the similarity between images based on the Euclidean distance between their representation vectors [8]....
[...]
TL;DR: A generic multi-layer and multi-modal framework for sports video analysis is proposed and several mid-level audio/visual features are introduced which are able to bridge the semantic gap between low-level features and high-level understanding.
Abstract: Advances in computing, networking, and multimedia technologies have led to a tremendous growth of sports video content and accelerated the need of analysis and understanding of sports video content. Sports video analysis has been a hot research area and a number of potential applications have been identified. In this paper, we summarize our research achievement on semantics extraction and automatic editorial content creation and adaptation in sports video analysis. We first propose a generic multi-layer and multi-modal framework for sports video analysis. Then we introduce several mid-level audio/visual features which are able to bridge the semantic gap between low-level features and high-level understanding. We also discuss emerging applications on editorial content creation and content enhancement/adaptation in sports video analysis, including event detection, sports MTV generation, automatic broadcast video generation, tactic analysis, player action recognition, virtual content insertion, and mobile sports video adaptation. Finally, we identify future directions in terms of research challenges remained and real applications expected.
36Â citations
"Feature based retrieval for animati..." refers background in this paper
...Changsheng Xu, Jian Cheng and Yi Zhang introduced a generic multi-layer and multimodal framework for sports video analysis which is able to bridge the semantic gap between low level features and high level understanding by adding several mid level audio/visual features [2]....
[...]
02 Jul 2007
TL;DR: It is demonstrated that increasing the number of concept detectors in a lexicon yields improved video retrieval performance indeed, and it is shown that combining concept detectors at query time has the potential to boost performance further.
Abstract: Until now, systematic studies on the effectiveness of concept detectors for video search have been carried out using less than 20 detectors, or in combination with other retrieval techniques. We investigate whether video search using just large concept detector lexicons is a viable alternative for present day approaches. We demonstrate that increasing the number of concept detectors in a lexicon yields improved video retrieval performance indeed. In addition, we show that combining concept detectors at query time has the potential to boost performance further. We obtain the experimental evidence on the automatic video search task of TRECVID 2005 using 363 machine learned concept detectors.
23Â citations
Additional excerpts
...Snoek and Marcel Worring [12] investigated whether video search using just large concept detector lexicon is a viable alternative for present day approach using two hypotheses....
[...]
...Cees G.M. Snoek and Marcel Worring [12] investigated whether video search using just large concept detector lexicon is a viable alternative for present day approach using two hypotheses....
[...]
IBM1
TL;DR: IBM Multimedia Search and Retrieval System is a Webbased technology that makes digital photos and video searchable through automated classification and indexing that automatically annotates images and videos by recognizing the semantic entities that are depicted in the content.
Abstract: IBM Multimedia Search and Retrieval System is a Webbased technology that makes digital photos and video searchable through automated classification and indexing [3] IBM system is unique in that it learns as it goes, helping users search immense multimedia content repositories faster and more effectively than ever before Marvel uses multi-modal machine learning techniques for bridging the semantic gap for multimedia content analysis and retrieval It automatically annotates images and videos by recognizing the semantic entities-such as scenes, objects, events, and people-that are depicted in the content
23Â citations
IBM1
TL;DR: This paper proposes an approach based on the use of knowledge models to building domain specific video information systems tailored to the domain of the data.
Abstract: Providing concept level access to video data requires, video management systems tailored to the domain of the data. Effective indexing and retrieval for high-level access mandates the use of domain knowledge. This paper proposes an approach based on the use of knowledge models to building domain specific video information systems. The key issues in such systems are identified and discussed.
22Â citations