Frame (networking)

About: Frame (networking) is a research topic. Over the lifetime, 142636 publications have been published within this topic receiving 937661 citations. The topic is also known as: data-link frame & data frame.

Good features to track

Abstract: No feature-based vision system can work unless good features can be identified and tracked from frame to frame. Although tracking itself is by and large a solved problem, selecting features that can be tracked well and correspond to physical points in the world is still hard. We propose a feature selection criterion that is optimal by construction because it is based on how the tracker works, and a feature monitoring method that can detect occlusions, disocclusions, and features that do not correspond to points in the world. These methods are based on a new tracking algorithm that extends previous Newton-Raphson style search methods to work under affine image transformations. We test performance with several simulations and experiments. >

Frame alignment processes, micromobilization, and movement participation.

Abstract: This paper attempts to further theoretical and empirical understanding of adherent and constituent mobilization by proposing and analyzing frame alignment as a conceptual bridge linking social psychological and resource mobilization views on movement participation. Extension of Goffinan's (1974) frame analytic perspective provides the conceptualltheoretical framework; field research on two religious movements, the peace movement, and several neighborhood movements provide the primary empirical base. Four frame alignment processes are identified and elaborated: frame bridging, frame amplification, frame extension, and frame transformation. The basic underlying premise is that frame alignment, of one variety or another, is a necessary condition for participation, whatever its nature or intensity, and that it is typically an interactional and ongoing accomplishment. The paper concludes with an elaboration of several sets of theoretical and research implications.

Tracking-Learning-Detection

Abstract: This paper investigates long-term tracking of unknown objects in a video stream. The object is defined by its location and extent in a single frame. In every frame that follows, the task is to determine the object's location and extent or indicate that the object is not present. We propose a novel tracking framework (TLD) that explicitly decomposes the long-term tracking task into tracking, learning, and detection. The tracker follows the object from frame to frame. The detector localizes all appearances that have been observed so far and corrects the tracker if necessary. The learning estimates the detector's errors and updates it to avoid these errors in the future. We study how to identify the detector's errors and learn from them. We develop a novel learning method (P-N learning) which estimates the errors by a pair of “experts”: (1) P-expert estimates missed detections, and (2) N-expert estimates false alarms. The learning process is modeled as a discrete dynamical system and the conditions under which the learning guarantees improvement are found. We describe our real-time implementation of the TLD framework and the P-N learning. We carry out an extensive quantitative evaluation which shows a significant improvement over state-of-the-art approaches.

Globally Consistent Range Scan Alignment for Environment Mapping

Abstract: A robot exploring an unknown environment may need to build a world model from sensor measurements. In order to integrate all the frames of sensor data, it is essential to align the data properly. An incremental approach has been typically used in the past, in which each local frame of data is aligned to a cumulative global model, and then merged to the model. Because different parts of the model are updated independently while there are errors in the registration, such an approach may result in an inconsistent model. In this paper, we study the problem of consistent registration of multiple frames of measurements (range scans), together with the related issues of representation and manipulation of spatial uncertainties. Our approach is to maintain all the local frames of data as well as the relative spatial relationships between local frames. These spatial relationships are modeled as random variables and are derived from matching pairwise scans or from odometry. Then we formulate a procedure based on the maximum likelihood criterion to optimally combine all the spatial relations. Consistency is achieved by using all the spatial relations as constraints to solve for the data frame poses simultaneously. Experiments with both simulated and real data will be presented.

Deep multi-scale video prediction beyond mean square error

Abstract: Learning to predict future images from a video sequence involves the construction of an internal representation that models the image evolution accurately, and therefore, to some degree, its content and dynamics. This is why pixel-space video prediction may be viewed as a promising avenue for unsupervised feature learning. In addition, while optical flow has been a very studied problem in computer vision for a long time, future frame prediction is rarely approached. Still, many vision applications could benefit from the knowledge of the next frames of videos, that does not require the complexity of tracking every pixel trajectories. In this work, we train a convolutional network to generate future frames given an input sequence. To deal with the inherently blurry predictions obtained from the standard Mean Squared Error (MSE) loss function, we propose three different and complementary feature learning strategies: a multi-scale architecture, an adversarial training method, and an image gradient difference loss function. We compare our predictions to different published results based on recurrent neural networks on the UCF101 dataset