摘要:
Systems and methods to generate a motion attention model of a video data sequence are described. In one aspect, a motion saliency map B is generated to precisely indicate motion attention areas for each frame in the video data sequence. The motion saliency maps are each based on intensity I, spatial coherence Cs, and temporal coherence Ct values. These values are extracted from each block or pixel in motion fields that are extracted from the video data sequence. Brightness values of detected motion attention areas in each frame are accumulated to generate, with respect to time, the motion attention model.
摘要:
Multi-label active learning may entail training a classifier with a set of training samples having multiple labels per sample. In an example embodiment, a method includes accepting a set of training samples, with the set of training samples having multiple respective samples that are each respectively associated with multiple labels. The set of training samples is analyzed to select a sample-label pair responsive to at least one error parameter. The selected sample-label pair is then submitted to an oracle for labeling.
摘要:
Systems and methods for smart media content thumbnail extraction are described. In one aspect program metadata is generated from recorded video content. The program metadata includes one or more key-frames from one or more corresponding shots. An objectively representative key-frame is identified from among the key-frames as a function of shot duration and frequency of appearance of key-frame content across multiple shots. The objectively representative key-frame is an image frame representative of the recorded video content. A thumbnail is created from the objectively representative key-frame.
摘要:
Kernelized spatial-contextual image classification is disclosed. One embodiment comprises generating a first spatial-contextual model to represent a first image, the first spatial-contextual model having a plurality of interconnected nodes arranged in a first pattern of connections with each node connected to at least one other node, generating a second spatial-contextual model to represent a second image using the first pattern of connections, and estimating the distance between corresponding nodes in the first spatial-contextual model and the second spatial-contextual model based on a relationship with adjacent connected nodes to determine a distance between the first image and the second image.
摘要:
Methods and systems for generic adaptive multimedia content delivery are described. In one embodiment, a novel framework features an abstract content model and an abstract adaptive delivery decision engine. The abstract content model recognizes important aspects of contents while hiding their physical details from other parts of the framework. The decision engine then makes content adaptation plans based on the abstracted model of the contents and needs little knowledge of any physical details of the actual contents. Thus, under the same framework, adaptive delivery of generic contents is possible.
摘要:
Various embodiments provide methods and systems for streaming data that can facilitate streaming during bandwidth fluctuations in a manner that can enhance the user experience. In one aspect, a forward-shifting technique is utilized to buffer data that is to be streamed, e.g. an enhancement layer in a FGS stream. Various techniques can drop layers actively when bandwidth is constant. The saved bandwidth can then be used to pre-stream enhancement layer portions. In another aspect, a content-aware decision can be made as to how to drop enhancement layers when bandwidth decreases. During periods of decreasing bandwidth, if a video segment does not contain important content, the enhancement layers will be dropped to keep the forward-shifting of the enhancement layer unchanged. If the enhancement layer does contain important content, it will be transmitted later when bandwidth increases.
摘要:
Correlative multi-label image annotation may entail annotating an image by indicating respective labels for respective concepts. In an example embodiment, a classifier is to annotate an image by implementing a labeling function that maps an input feature space and a label space to a combination feature vector. The combination feature vector models both features of individual ones of the concepts and correlations among the concepts.
摘要:
Systems and methods for automatic generation of a browsing path across image content to present areas with high attention value are described. In particular, an image is modeled via multiple visual attentions to create a respective set of attention objects for each modeled attention. The attention objects and their respective attributes are analyzed to generate a browsing path to select ones of the attention objects. The browsing path is generated to optimize the rate of information gain from the attention objects as a function of information unit cost in terms of time constraints associated with multiple image browsing modes.
摘要:
Systems and methods for image attention analysis are described. In one aspect, image attention is modeled by preprocessing an image to generate a quantized set of image blocks. A contrast-based saliency map for modeling one-to-three levels of image attention is then generated from the quantized image blocks.
摘要:
A multimedia object retrieval and annotation system integrates an annotation process with object retrieval and relevance feedback processes. The annotation process annotates multimedia objects, such as digital images, with semantically relevant keywords. The annotation process is performed in background, hidden from the user, as the user conducts normal searches. The annotation process is “semi-automatic” in that it utilizes both keyword-based information retrieval and content-based image retrieval techniques to automatically search for multimedia objects, and then encourages users to provide feedback on the retrieved objects. The user identifies objects as either relevant or irrelevant to the query keywords and based on this feedback, the system automatically annotates the objects with semantically relevant keywords and/or updates associations between the keywords and objects. As the retrieval-feedback-annotation cycle is repeated, the annotation coverage and accuracy of future searches continues to improve.