Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
44,631
result(s) for
"Video data"
Sort by:
Killer tapes and shattered screens
Since the mid-1980s, US audiences have watched the majority of movies they see on a video platform, be it VHS, DVD, Blu-ray, Video On Demand, or streaming media. Annual video revenues have exceeded box office returns for over twenty-five years. In short, video has become the structuring discourse of US movie culture. Killer Tapes and Shattered Screens examines how prerecorded video reframes the premises and promises of motion picture spectatorship. But instead of offering a history of video technology or reception, Caetlin Benson-Allott analyzes how the movies themselves understand and represent the symbiosis of platform and spectator. Through case studies and close readings that blend industry history with apparatus theory, psychoanalysis with platform studies, and production history with postmodern philosophy, Killer Tapes and Shattered Screens unearths a genealogy of post-cinematic spectatorship in horror movies, thrillers, and other exploitation genres. From Night of the Living Dead (1968) through Paranormal Activity (2009), these movies pursue their spectator from one platform to another, adapting to suit new exhibition norms and cultural concerns in the evolution of the video subject.
eBook
SRFCNM: Spatiotemporal recurrent fully convolutional network model for salient object detection
Video saliency detection has recently been widely used because of its ability to distinguish significant regions of interest. It has several applications, such as video segmentation, abnormal activity detection, video summarization, etc. This research paper develops a novel technique for video saliency detection known as Spatiotemporal Recurrent Fully Convolutional Network Model (SRFCNM). This model uses recurrent convolutional layers to represent spatial and temporal features of superpixels for element uniqueness. The model is trained in two phases; initially, we pre-train the model on the segmented data sets and then fine-tune it for saliency detection, which allows the network to learn salient objects more accurately. The uniqueness of integrating saliency maps with recurrent convolutional layers and spatiotemporal characteristics facilitates the robust representation of salient objects to capture the relevant features. The SRFCNM model is extensively estimated on the challenging datasets viz. SegTrackV2, FBMS and DAVIS. Our model is compared with the existing Deep Learning and Convolutional Neural Network algorithms. This research demonstrates that SRFCNM outperforms the state-of-the-art saliency models considerably over the three public datasets regarding accuracy recall and mean absolute error (MAE). The proposed SRFCNM model produces the lowest MAE values, 3.2%, 3.5%, and 7.5%, for SegTrackV2, DAVIS, and FBMS datasets, respectively, with hand-crafted color features, compared with the existing models.
Journal Article
Intelligent video surveillance systems : an algorithmic approach
This book will provide an overview of techniques for visual monitoring including video surveillance and human activity understanding. It will present the basic techniques of processing video from static cameras, starting with object detection and tracking. The author will introduce further video analytic modules including face detection, trajectory analysis and object classification. Examining system design and specific problems in visual surveillance, such as the use of multiple cameras and moving cameras, the author will elaborate on privacy issues focusing on approaches where automatic processing can help protect privacy-- Provided by publisher.
Book
Generalized zero-shot learning for action recognition with web-scale video data
Action recognition in surveillance video makes our life safer by detecting the criminal events or predicting violent emergencies. However, efficient action recognition is not free of difficulty. First, there are so many action classes in daily life that we cannot pre-define all possible action classes beforehand. Moreover, it is very hard to collect real-word videos for certain particular actions such as steal and street fight due to legal restrictions and privacy protection. These challenges make existing data-driven recognition methods insufficient to attain desired performance. Zero-shot learning is potential to be applied to solve these issues since it can perform classification without positive example. Nevertheless, current zero-shot learning algorithms have been studied under the unreasonable setting where seen classes are absent during the testing phase. Motivated by this, we study the task of action recognition in surveillance video under a more realistic generalized zero-shot setting, where testing data contains both seen and unseen classes. To our best knowledge, this is one of the first works to study video action recognition under the generalized zero-shot setting. We firstly perform extensive empirical studies on several existing zero-shot leaning approaches under this new setting on a web-scale video data. Our experimental results demonstrate that, under the generalize setting, typical zero-shot learning methods are no longer effective for the dataset we applied. Then, we propose to deploy generalized zero-shot learning which transfers the knowledge of Web video to detect the anomalous actions in surveillance videos. To verify the effectiveness of methods, we further construct a new surveillance video dataset consisting of nine action classes related to the public safety situation.
Journal Article
Performance of Agency in Real-Life Encounters
This article explores the performance of agency within the context of unequal power resources and structural constraint. Based on 23 video-recorded placement meetings in three homeless shelters, we find that participants’ agency is the outcome of both collaboration and resistance. To avoid interaction that fails to empower, social actors engage in “repair work” and face-saving practices. When clients display “wrong face”—that is, bring problems to the table that are not considered “reasonable”—then the service providers engage in “repair work.” Participants turn conflict into collaborative agency because interactions that fail to deliver mutually empowering forms of agency have costs for both: clients’ problems are not solved, and service providers fail to reach organizational goals.
Journal Article
YouTube, Google, Facebook: 21st Century Online Video Research and Research Ethics
Since the early 2000s, the proliferation of cameras in devices such as mobile phones, closed-circuit television (CCTV), or body cameras has led to a sharp increase in video recordings of human interaction and behavior. Through websites that employ user-generated content (e.g., YouTube) and live streaming sites (e.g., GeoCam), access to such videos virtually is at the fingertips of social science researchers. Online video data offer great potential for social science research to study an array of human interaction and behavior, but they also raise ethical questions to which existing guidelines and publications only provide partial answers. In our article we address this gap, drawing on existing ethical discussions and applying them to the use of online video data. We examine five areas in which online video research raises specific questions or promises unique potentials: informed consent, analytic opportunities, privacy, transparency, and minimizing harm to participants. We discuss their interplay and how these areas can inform practitioners, reviewers, and interested readers of online video studies when evaluating the ethical standing of a study. With this study, we contribute to an informed and transparent discussion about ethics in online video research.
Journal Article