Effective Content-based Video Retrieval Research Articles

Learning Unsupervised Visual Representations using 3D Convolutional Autoencoder with Temporal Contrastive Modeling for Video Retrieval

The rapid growth of tag-free user-generated videos (on the Internet), surgical recorded videos, and surveillance videos has necessitated the need for effective content-based video retrieval systems. Earlier methods for video representations are based on hand-crafted, which hardly performed well on the video retrieval tasks. Subsequently, deep learning methods have successfully demonstrated their effectiveness in both image and video-related tasks, but at the cost of creating massively labeled datasets. Thus, the economic solution is to use freely available unlabeled web videos for representation learning. In this regard, most of the recently developed methods are based on solving a single pretext task using 2D or 3D convolutional network. However, this paper designs and studies a 3D convolutional autoencoder (3D-CAE) for video representation learning (since it does not require labels). Further, this paper proposes a new unsupervised video feature learning method based on joint learning of past and future prediction using 3D-CAE with temporal contrastive learning. The experiments are conducted on UCF-101 and HMDB-51 datasets, where the proposed approach achieves better retrieval performance than state-of-the-art. In the ablation study, the action recognition task is performed by fine-tuning the unsupervised pre-trained model where it outperforms other methods, which further confirms the superiority of our method in learning underlying features. Such an unsupervised representation learning approach could also benefit the medical domain, where it is expensive to create large label datasets.

An efficient access method for multimodal video retrieval

This paper presents the Slim 2 -tree, an efficient and effective content-based video retrieval technique allowing the use of multiple modalities within a single index structure. Slim 2 -tree is capable of dealing with different distance measures for the modalities and can perform both multimodal and unimodal searches using the same tree structure. Experimental studies on a large real dataset show the video similarity search performance of the proposed technique. Additionally, we present experiments comparing our method against state-of-the-art of multimodal solutions. Comparative test results demonstrate that our technique improves the performance of video similarity queries.

A Content Based Video Retrieval Analysis System with Extensive Features by Using Kullback-Leibler

AbstractContent-based video retrieval systems have shown great potential in supporting decision making in clinical activities, teaching, and biological research. In content-based video retrieval, feature combination plays a key role. As a result content-based retrieval of all different type video data turns out to be a challenging and vigorous problem. This paper presents an effective content based video retrieval system, which recognizes and retrieves videos with three different types of visual effects. The raw video information is divided into shots and also the object feature, movement feature and also the occlusion options are extracted from these shots and also the feature library is used for the storage method of those options. Advanced on, the Kullback-Leibler distance is computed among the options of the feature library and also the options of the question clip that's extracted within the similar manner. The results show that it is possible to improve a system for content-based video retrieval by ...

Kullback-Leibler similarity measures for effective content based video retrieval

Recent advancements in the multimedia technologies allow the capture and storage of video data with relatively inexpensive computers. As the necessity to query these data competently becomes significant, the amount of broadly accessible video data grows. As a result, content-based retrieval of video data turns out to be a demanding and vital problem. In this paper, an effective content-based video retrieval system is proposed. The raw video data are segmented into shots and the object feature, movement feature and the occlusion feature are extracted from these shots and the feature library is utilised for the storage process of these features. Subsequently, the Kullback–Leibler distance is computed among the features of the feature library and the features of the query clip which is extracted in the similar manner. Hence, with the aid of the Kullback–Leibler distance, the similar videos are extracted from the collection of videos based on the given query video clip in an effective manner.

The retrieval of motion event by associations of temporal frequent pattern growth

With the development of Internet technology, a vast number of video data are available. Mining the hidden relationship among semantic concepts in video is important for effective content-based video retrieval and has gained great attention recently. Cloud computing, as a cost-effective solution, has become popular in mining video data for storing the distributed data and computations. In this paper, we have developed a novel method based on frequent pattern tree (FPTree) for mining association rules in video retrieval. The core of the method is to extend the structure of FPTree by temporal parameter in motion events. Firstly, we get semantic concepts based trajectory retrieval, Ncuts has been used to classify the sub-events by trajectory segmentation, and the sub-events in each event have been annotated. Secondly, the new modeling, called temporal frequent pattern tree (TFPTree) is used to store motion event semantic concepts. And we propose the TFP-Growth algorithm to mine temporal frequent patterns from TFPTree for finding the rules of the motion events. As video datasets grow large, cloud-based infrastructure has been used to support our computing. The experiment shows our method is both efficient and effective in improving the accuracy of semantic concept detection in video retrieval.

A Comprhensive CBVR System Based on Spatiotemporal Features Such as Motion,Quantized Color and Edge Density Features

Rapid develop ment of the mult imedia and the associated technologies urge the processing of a huge database of video clips. The processing efficiency depends on the search methodologies utilized in the video processing system. Use of inappropriate search methodologies may make the processing system ineffective. Hence, an effective video retrieval system is an essential pre-requisite for searching relevant videos from a huge collection of videos. In this paper, an effective content based video retrieval system based on some dominant features such as motion, color and edge is proposed. The system is evaluated using the video clips of format MPEG-2 and then precision-recall is determined for the test clip.

An Effective CBVR System Based on Motion, Quantized Color and Edge Density Features

Rapid development of the multimedia and the associated technologies urge the processing of a huge database of video clips. The processing efficiency lies on the search methodologies utilized in the video processing system. Usage of inappropriate search methodologies may make the processing system ineffective. Hence, an effective video retrieval system is an essential pre-requisite for searching relevant videos from a huge collection of videos. In this paper, an effective content based video retrieval system based on some dominant features such as motion, color and edge is proposed. The system is evaluated using the video clips of format MPEG-2 and then precision-recall is determined for the test clip.

Effective content-based video retrieval using pattern-indexing and matching techniques

Recently, multimedia data grows rapidly due to the advanced multimedia capturing devices, such as digital video recorder, mobile camera and so on. Since conventional query-by-text retrieval cannot satisfy users’ requirements in finding the desired videos effectively, content-based video retrieval is regarded as one of the most practical solutions to improve the retrieval quality. In addition, video retrieval using query-by-image is not successful in associating the videos with user’s interest either. In this paper, we propose an innovative method to achieve the high quality of content-based video retrieval by discovering the temporal patterns in the video contents. On basis of the discovered temporal patterns, an efficient indexing technique and an effective sequence matching technique are integrated to reduce the computation cost and to raise the retrieval accuracy, respectively. Experimental results reveal that our approach is very promising in enhancing content-based video retrieval in terms of efficiency and effectiveness.

Association and Temporal Rule Mining for Post-Filtering of Semantic Concept Detection in Video

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Automatic semantic concept detection in video is important for effective content-based video retrieval and mining and has gained great attention recently. In this paper, we propose a general post-filtering framework to enhance robustness and accuracy of semantic concept detection using association and temporal analysis for concept knowledge discovery. Co-occurrence of several semantic concepts could imply the presence of other concepts. We use association mining techniques to discover such inter-concept association relationships from annotations. With discovered concept association rules, we propose a strategy to combine associated concept classifiers to improve detection accuracy. In addition, because video is often visually smooth and semantically coherent, detection results from temporally adjacent shots could be used for the detection of the current shot. We propose temporal filter designs for inter-shot temporal dependency mining to further improve detection accuracy. Experiments on the TRECVID 2005 dataset show our post-filtering framework is both efficient and effective in improving the accuracy of semantic concept detection in video. Furthermore, it is easy to integrate our framework with existing classifiers to boost their performance. </para>