- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - Only-Reference Video Quality Assessment for Video Coding Using Convolutional Neural Network
摘要： Conventional video quality assessment methods are either full-, reduced-, or no-reference methods that need to access decoded videos. Hence, to calculate quality of decoded video in video coding regarding an image/video quality metric, complete encoding and decoding have to executed, which is computationally expensive. To address this problem, we propose to estimate quality of decoded videos from the original video only (i.e., only-reference) using convolutional neural network, as if the original video is encoded using a range of quantization parameter. The proposed network is shallow and can be trained to estimate various video quality metrics. Furthermore, among potential rate control applications using the proposed network, we demonstrate achieving a targeted decoded-video quality by selecting a proper quantization parameter before actually encoding.
关键词： only-reference,Video quality assessment,convolutional neural network,video coding
更新于2025-09-09 09:28:46
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - Video Summarization via Weighted Neighborhood Based Representation
摘要： The recent explosive growth of multimedia data has posed a new set of challenges in computer vision, and video summarization (VS) techniques are increasingly important to automatically summarize a large amount of multimedia data in an effective and efficient manner. Recent years have witnessed the rise and developments of sparse representation based approaches for VS. While the existing methods select keyframes according to the information contained in the single frame, and such a selection based solely on single-frame information may not be robust. Therefore, in this paper, the information of the single frame’s neighborhood is taken into consideration, and different weights are assigned to these neighbouring frames. We formulate the VS problem as a weighted neighborhood based representation model, and design a greedy pursuit algorithm to extract keyframes. Experimental results on a benchmark dataset demonstrate that the proposed method can outperform the state of the arts.
关键词： sparse representation,weighted neighborhood,Video summarization
更新于2025-09-09 09:28:46
-
LiDAR Validation of a Video-Derived Beachface Topography on a Tidal Flat
摘要： Increasingly used shore-based video stations enable a high spatiotemporal frequency analysis of shoreline migration. Shoreline detection techniques combined with hydrodynamic conditions enable the creation of digital elevation models (DEMs). However, shoreline elevations are often estimated based on nearshore process empirical equations leading to uncertainties in video-based topography. To achieve high DEM correspondence between both techniques, we assessed video-derived DEMs against LiDAR surveys during low energy conditions. A newly installed video system on a tidal flat in the St. Lawrence Estuary, Atlantic Canada, served as a test case. Shorelines were automatically detected from time-averaged (TIMEX) images using color ratios in low energy conditions synchronously with mobile terrestrial LiDAR during two different surveys. Hydrodynamic (waves and tides) data were recorded in-situ, and established two different cases of water elevation models as a basis for shoreline elevations. DEMs were created and tested against LiDAR. Statistical analysis of shoreline elevations and migrations were made, and morphological variability was assessed between both surveys. Results indicate that the best shoreline elevation model includes both the significant wave height and the mean water level. Low energy conditions and in-situ hydrodynamic measurements made it possible to produce video-derived DEMs virtually as accurate as a LiDAR product, and therefore make an effective tool for coastal managers.
关键词： erosion,beach morphology,video monitoring,Atlantic Canada,mobile terrestrial LiDAR,shoreline detection
更新于2025-09-09 09:28:46
-
[ACM Press the 2nd International Conference - Las Vegas, NV, USA (2018.08.27-2018.08.29)] Proceedings of the 2nd International Conference on Vision, Image and Signal Processing - ICVISP 2018 - Perceptually Lossless Video Compression with Error Concealment
摘要： We present a video compression framework that has several components. First, we aim at achieving perceptually lossless compression. Several well-known video codecs in the literature have been evaluated and the performance was assessed using several well-known performance metrics. Second, we investigated the impact of error concealment algorithms for handling corrupted pixels due to transmission errors in communication channels. Extensive experiments using actual videos have been performed to demonstrate the proposed framework.
关键词： Perceptually Lossless,Video Compression,Error Concealment
更新于2025-09-09 09:28:46
-
[IEEE 2018 International Conference on Smart City and Emerging Technology (ICSCET) - Mumbai, India (2018.1.5-2018.1.5)] 2018 International Conference on Smart City and Emerging Technology (ICSCET) - Tracking People In Real Time Video Footage Using Facial Recognition
摘要： As of now, there is very less knowledge and use of Facial Recognition System for security surveillance in India. This project proposes a system which will use Facial Recognition to track or search a target person from a real time video feed, like a video feed from a surveillance system. Firstly, the system is provided with a Live Video footage of the area that has to be scanned. Then it is provided with an input data set of images of a targeted person for example, a missing person, criminal, etc. Once the input is provided the system will extract a predefined set of facial characteristics from the Input Dataset and create a training module which will help in searching the person from the real-time video footage. If a match is found, the system will identify and mark the person. Also one of the main objectives of this project is to develop the above-mentioned system in conjunction with the existing Surveillance system i.e. to make it compatible with the already installed surveillance cameras in- order keep the costs and hassle of running it at a minimum. The applications of this proposed system can be in Government Organizations like Police, Military, Municipal Corporations, Large Companies, etc for tracking people.
关键词： Facial Recognition,People Tracking,Intruder Detection.,Facial Detection,Real Time Video
更新于2025-09-09 09:28:46
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - A New Foreground Segmentation Method for Video Analysis in Different Color Spaces
摘要： A new foreground segmentation method is presented in this paper for video analysis. Specifically, a new feature representation scheme is first proposed in different color spaces, namely, the RGB, the YIQ, and the YCbCr color spaces. The new feature vector, which integrates the color values in a particular color space, the horizontal and vertical Haar wavelet features, and the temporal difference features, enhances the discriminatory power. A new Global Foreground Modeling (GFM) method is then presented to improve upon the popular video analysis approaches. The Bayes classifier is finally applied for foreground segmentation in video. Experimental results using the New Jersey Department of Transportation (NJDOT) traffic video sequences show that the new foreground segmentation method achieves better performance than the popular video analysis methods.
关键词： Global Foreground Modeling,video analysis,Haar wavelet features,Bayes classifier,temporal difference features,foreground segmentation,color spaces
更新于2025-09-09 09:28:46
-
12.2: A 3D Display Parallel System: Light Field Re-rendering and Depth Sense Optimization
摘要： The main benefit of 3D display over 2D display is the obvious ability to create a more lifelike character with high depth sense. However, the limitation of human eye’s visual mechanism, unartful 3D scene structure design, or bad viewing condition always emerges poor depth perception experience or even physiological discomfort during the watching time, which is often sub-optimal for mass high-quality 3D display productions. To solve this problem, we propose a novel 3D display parallel system for depth sense optimization and it empirically guides how the light field should be re-rendered. Structurally, the parallel system consists of an artificial perception measurement system, a display evaluation model and a light field display rendering system, which includes the display calibration, scene capture, light field data processing and display. Particularly, the system can systematically analyze and model various factors affecting the depth sense which learned through the measurement system, like scene structure, objects’ speeds in 3D video and so on. And those sense factors can be personally modified or increased according to the viewer’s demands or technical improvement. Moreover, the light field could be real-time re-rendered, based on some image processing technology, optical flow analysis and object segmentation (or tracking) (especially the one-shot video segmentation). Theory and algorithms are developed and experimental validation results show a superior performance.
关键词： Light Field Re-rendering,Depth Sense,3D Video Processing,One-shot Segmentation,3D Display Parallel System
更新于2025-09-09 09:28:46
-
[IEEE 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Krakow, Poland (2018.10.16-2018.10.18)] 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Virtual Video Synthesis for Personalized Training
摘要： Online personal training allows users to work out from the comfort of their own homes using workout videos designed by fitness instructors. Users of such applications can use their device (PC, laptop, smart TV, etc.) camera and work out with others in a group setting, enabling plethora of intertwined benefits. In order to enhance training efficiency, it could be helpful for the trainee to superimpose his/her human silhouette, giving the differences of his/her exercise over the trainer’s movements. One way to proceed towards this direction is to have a camera recording the video of the trainee during the exercise, which should be presented in contrast to the instructor’s video on the device screen. In this work, we explore this direction and present traditional background estimation approaches in combination with foreground extraction techniques using videos recorded with static cameras. It is shown that none of the presented methods is able to efficiently face all possible challenges, like slow moving object (foreground) or presence of the moving object at the phase of background initialization, problems that mainly appear in in yoga exercise. As an alternative, we propose a series of techniques including an initial background reconstruction method followed by a selective updating scheme. In this way, the background image adaptively converges to the ground truth data enabled by the merging of from detected moving regions (temporal processing) and color-based regions (spatial processing) of the video segment. Finally, we also apply the proposed method in space surveillance applications, using surveillance cameras, in order to evaluate the generality and efficiency of the proposed approach.
关键词： image background reconstruction,silhouette extraction,fusion of temporal and spatial information,motion tracking,video processing
更新于2025-09-09 09:28:46
-
[IEEE 2018 IEEE Broadcast Symposium (BTS) - Arlington, VA, USA (2018.10.9-2018.10.11)] 2018 IEEE Broadcast Symposium (BTS) - 4K HDR Workflow: from Capture to Display
摘要： In the past several years, TV picture quality has improved with the introduction of 4K and wide color gamut. In addition, recently HDR (High Dynamic Range) has nurtured the picture performance evolution. Therefore TV images have become much more captivating. But, HDR seems difficult to understand correctly, since it contains a new concept, several different technologies, and standards from the status quo. In this paper, the meaning, benefit, and ecosystem of HDR will be explained. How HDR affects and what HDR requires of the system of TV/display will be explained for furthering one's understanding and utilization of HDR.
关键词： HLG,EOTF,HDR,video,HDR10
更新于2025-09-09 09:28:46
-
[IEEE 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Krakow, Poland (2018.10.16-2018.10.18)] 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Power Transmission Lines Inspection using Properly Equipped Unmanned Aerial Vehicle (UAV)
摘要： The inspection of power transmission lines is an important task that enhances the reliability of Electricity Distribution Network Operators. This task can be performed in a low-cost way using unmanned aircrafts. At the present study, we examine the effectiveness of using basic image processing methods on image data of the power lines acquired by an unmanned aerial vehicle (UAV). The specific UAV was assembled for the present work under the considerations that arise from the purpose of the inspection of power transmission lines. Two methodologies are proposed differing on the pre- processing required in order to detect the location of the lines on the video images. Both proposed methodologies were tested in real-world cases, with the image background in each case to be characterized of non-uniform texture, i.e. the natural terrain is rugged at some locations, wooded land at some other or it is road that appears at the same hue as the aerial power lines. We examined the case of a broken line where the methodologies result in successful detection of the power lines before and after the discontinuity of the power line. The proposed work offers a robust and low-cost way for the inspection of power transmission lines and so an effective way to detect the location where a cable fault has occurred.
关键词： Hough transform,aerial video processing,UAV remote sensing,power line inspection,parametric training
更新于2025-09-09 09:28:46