Original Video Sequence Research Articles

Sliding Adjustment for 3D Video Representation

This paper deals with video coding of static scenes viewed by a moving camera. We propose an automatic way to encode such video sequences using several 3D models. Contrary to prior art in model-based coding where 3D models have to be known, the 3D models are automatically computed from the original video sequence. We show that several independent 3D models provide the same functionalities as one single 3D model, and avoid some drawbacks of the previous approaches. To achieve this goal we propose a novel algorithm of sliding adjustment, which ensures consistency of successive 3D models. The paper presents a method to automatically extract the set of 3D models and associate camera positions. The obtained representation can be used for reconstructing the original sequence, or virtual ones. It also enables 3D functionalities such as synthetic object insertion, lightning modification, or stereoscopic visualization. Results on real video sequences are presented.

Categorization of visualization tools in aspects of chemical research and education

AbstractChemistry is an experimental science whose research subjects are often not directly available within the close dimensions of a time or space. To optimize a particular teaching process or research, it is necessary to visualize phenomena, processes, or objects. Consequently, visualization is significant to problem solving in research and teaching. Visualization is applicable in those situations when it is required to reduce the dimensions of experimental objects to given numbers and values accessible to direct experience. In educational practice, this is related to one of the following situations: (1) the experiment is too long or too short; (2) the dimensions of the examined object are too small or too large; (3) the environment of the experiment is not accessible; (4) the parameters of the experiment or its effects are not directly available to the observer's senses; (5) there is a need for multiple revisions of the experiment; (6) the experiment is difficult to arrange or revise effectively; (7) the experiment is dangerous; (8) the experiment is too expensive; etc. The presentation of models of physicochemical processes and phenomena, models of atoms and molecules, and mechanisms of reaction, as well as experimental data and their mathematical models, provides a basis for considering the formulation of correct conclusions. The appropriate choice of visualization algorithms is decisive of the effectiveness of theoretical analyses. The main aspect of visualization is obviously its application to teaching or research. Thus, the choice of techniques depends on whether it is used for recognition, explanation, memorization, inspiration, or motivation. In research, the importance as well as area of usefulness of three main categories of visualization have also been established. These categories are: illustration—relating to the presentation of an existing picture of object; representation—the presentation of an image assigned to an object; and visualization—related to the presentation of an image created for the object. Research work on preparing means for visualization confirms that, depending on the particular aspect of categorizing visualization, it may be useful, as well, to assign additional criteria such as dimensions, degree, functions, methodological contexts, experimental relationship, and medial effectiveness into visualization. In recent years, at the Institute of Chemical Education of Poznan's Adam Mickiewicz University such research was undertaken. Its main aim was to determine the educational effectiveness of different types of chemical visualizations. Methods for the visualization of chemical issues were developed and a series of original video and animation sequences presenting different aspects of chemical visualization prepared. Interactive Internet instructions sets were produced and methods of visualization categorized. Effectiveness in using different types of chemical visualizations was estimated. © 2002 Wiley Periodicals, Inc. Int J Quantum Chem, 2002

Event-Driven Video Abstraction and Visualization

In this paper, we propose a new video summarization procedure that produces a dynamic (video) abstract of the original video sequence. Our technique compactly summarizes a video data by preserving its original temporal characteristics (visual activity) and semantically essential information. It relies on an adaptive nonlinear sampling. The local sampling rate is directly proportional to the amount of visual activity in localized sub-shot units of the video. To get very short, yet semantically meaningful summaries, we also present an event-oriented abstraction scheme, in which two semantic eventss emotional dialogue and violent action, are characterized and abstracted into the video summary before all other events. If the length of the summary permits, other non key events are then added. The resulting video abstract is highly compact.

A new multi-level statistical model for variable bit rate MPEG sources over ATM networks and its performance study

In this paper, we propose a new traffic model for MPEG-coded video sequences. The proposed modeling scheme uses scene-based traffic characteristics and considers the correlations between frames of consecutive group of pictures (GOPs). Using a simple scene detection algorithm, scene changes are modeled by a state transition matrix and the number of GOPs of a scene state is modeled by a geometric distribution. Frames of a scene are modeled by the mean I, P, and B frame sizes of each state. For more accurate traffic modeling, the residual bits that represent the difference between the original frame size and the mean frame size of each frame type are compensated by autoregressive processes. The modeling results show that our scene-based model can capture the statistical traffic characteristics of the original video sequences well and estimate the queueing performance with good approximation quality.

Video quality and traffic QoS in learning-based subsampled and receiver-interpolated video sequences

Sources of real-time traffic are generally highly unpredictable with respect to the instantaneous and average load which they create. Yet such sources will provide a significant portion of traffic in future networks, and will significantly affect the overall performance of and quality of service. Clearly high levels of compression are desirable as long as video quality remains satisfactory, and our research addresses this key issue with a novel learning-based approach. We propose the use of neural networks (NNs) as post-processors for any existing video compression scheme. The approach is to interpolate video sequences and compensate for frames which may have been lost or deliberately dropped. We show that deliberately dropping frames will significantly reduce the amount of offered traffic in the network, and hence the cell loss probability and network congestion, while the NN post-processor will preserve most of the desired video quality. Dropping frames at the sender or in the network is also a fast way to react to network overload and reduce congestion. Our interpolation techniques at the receiver, including neural network-based algorithms, provide output frame rates which are identical to (or possibly higher than) the original video sequence's frame rate. The resulting video quality is essentially equivalent to the sequence without frame drops, despite the loss of a significant fraction of the frames. Experimental evaluation using real video sequences is provided or interpolation with a connectionist NN using the backpropagation learning algorithm, the random NN (RNN) in a feed-forward configuration with its associated learning algorithm, and cubic spline interpolation. The experiments show that when more frames are being dropped or lost, the RNN performs generally better than the other techniques in terms of resulting video quality and overall performance. When the fraction of dropped frames is small, cubic splines offer better performance. Experimental data shows that this receiver-reconstructed subsampling technique significantly reduces the cell loss rates in an asynchronous transfer mode switch for different buffer sizes and service rates.

Capture and representation of human walking in live video sequences

Extracting human representations from video has vast applications. In this paper, we present a knowledge-based framework to capture metarepresentations for real-life video with human walkers. The system models the human body as an articulated object and the human walking as a cyclic activity with highly correlated temporal patterns. We extract for each of the body parts its motion, shape, and texture. Once available, this structural information can be used to manipulate or synthesize the original video sequence, or animate the walker with a different motion in a new synthesized video.

Falls On The Track Detection In Public Transport By Artificial Vision

The aim of this application is to improve safety by detecting automatically people or objects falling on or crossing the in public transport. Through computer vision processing, the system to be developed determines and sends, in due time and via a satisfactory Human Machine Interface (HMI), alarms and the corresponding video image to the systems operators. In the current version of the system, to detect falls on the tracks, we propose to fix the cameras on the ceiling, aimed vertically so that the framing delineates clearly what is standing on the platform from what is above the tracks. To detect falls, only the moving edges in the tracks part of the picture will trigger a warning. With this particular camera position, passengers bent over the edge are detected as possible incidents and potential accidents, though other camera placements do not detect such incident scenarios and are being considered with the French operator RATP. According to the very hard working conditions (-uncontrolled lightening, changes of background, unknown moving objects) a specific robust algorithm has been designed to perform the detection of the moving edges. This algorithm is based on the analysis of the differences of gray level between successive frames in an original video sequence of images. Edge extraction, performed on the difference of successive images, allows to retain only the edges of moving objects in the frame. Transactions on the Built Environment vol 34, © 1998 WIT Press, www.witpress.com, ISSN 1743-3509

Digital watermarking for copyright protection of MPEG2 compressed video

Digital watermarking is the technique which embeds an invisible signal including owner identification and copy control information into multimedia data such as audio, video, images, for copyright protection. This paper proposes a new watermarking technique which is appropriate for the MPEG-2 video coding system by extending the direct sequence spread spectrum. The proposed watermarking scheme in this paper minimizes the perceptual degradation of video quality, caused by the embedded watermark through controlling the strength and area of embedding with respect to the global and local characteristics of the video sequences. And this technique can also directly extract the embedded watermark from the watermarked MPEG-2 compressed video stream without using original video sequences. Simulation results show that the proposed scheme is robust to various attacks to remove the embedded watermark and is feasible for real time system implementation.