Recovery and Reasoning About Occlusions in 3D Using Few Cameras with Applications to 3D Tracking

In this work we propose algorithms to learn the locations of static occlusions and reason about both static and dynamic occlusion scenarios in multi-camera scenes for 3D surveillance (e.g., reconstruction, tracking). We will show that this leads to a computer system which is able to more effectively track (follow) objects in video when they are obstructed from some of the views. Because of the nature of the application area, our algorithm will be under the constraints of using few cameras (no more than 3) that are configured wide-baseline. Our algorithm consists of a learning phase, where a 3D probabilistic model of occlusions is estimated per-voxel, per-view over time via an iterative framework. In this framework, at each frame the visual hull of each foreground object (person) is computed via a Markov Random Field that integrates the occlusion model. The model is then updated at each frame using this solution, providing an iterative process that can accurately estimate the occlusion model over time and overcome the few-camera constraint. We demonstrate the application of such a model to a number of areas, including visual hull reconstruction, the reconstruction of the occluding structures themselves, and 3D tracking.     

3D Video Billboard Clouds

3D video billboard clouds reconstruct and represent a dynamic three-dimensional scene using displacement-mapped billboards. They consist of geometric proxy planes augmented with detailed displacement maps and combine the generality of geometry-based 3D video with the regularization properties of image-based 3D video. 3D video billboards are an image-based representation placed in the disparity space of the acquisition cameras and thus provide a regular sampling of the scene with a uniform error model. We propose a general geometry filtering framework which generates time-coherent models and removes reconstruction and quantization noise as well as calibration errors. This replaces the complex and time-consuming sub-pixel matching process in stereo reconstruction with a bilateral filter. Rendering is performed using a GPU-accelerated algorithm which generates consistent view-dependent geometry and textures for each individual frame. In addition, we present a semi-automatic approach for modeling dynamic three-dimensional scenes with a set of multiple 3D video billboards clouds.     

3D face and motion estimation from sparse points using adaptive bracketed minimization

This paper presents a novel method for estimating camera motion and reconstructing human face from a video sequence. The coarse-to-fine method is applied via combining the concepts of Powell’s minimization with gradient descent. Sparse points defining the human face in every frame are tracked using the active appearance model. The case of occluded points, even for self-occlusion, does not pose a problem in the proposed method. Robustness in the presence of noise and 3D accuracy using this method is also demonstrated. Examples of face reconstruction using other methods including trifocal tensor, Powell’s minimization, and gradient descent are also compared to the proposed method. Experiments on both synthetic and real faces are presented and analyzed. Also, different camera movement paths are illustrated. All real-world experiments used an off-the-shelf digital camera carried by a human walking without using any dolly to demonstrate the robustness and practicality of the proposed method.     

Dense 3D face alignment from 2D video for real-time use

To enable real-time, person-independent 3D registration from 2D video, we developed a 3D cascade regression approach in which facial landmarks remain invariant across pose over a range of approximately 60°. From a single 2D image of a person's face, a dense 3D shape is registered in real time for each frame. The algorithm utilizes a fast cascade regression framework trained on high-resolution 3D face-scans of posed and spontaneous emotion expression. The algorithm first estimates the location of a dense set of landmarks and their visibility, then reconstructs face shapes by fitting a part-based 3D model. Because no assumptions are required about illumination or surface properties, the method can be applied to a wide range of imaging conditions that include 2D video and uncalibrated multi-view video. The method has been validated in a battery of experiments that evaluate its precision of 3D reconstruction, extension to multi-view reconstruction, temporal integration for videos and 3D head-pose estimation. Experimental findings strongly support the validity of real-time, 3D registration and reconstruction from 2D video. The software is available online at http://zface.org.     

Combining 3D flow fields with silhouette-based human motion capture for immersive video

In recent years, the convergence of computer vision and computer graphics has put forth a new field of research that focuses on the reconstruction of real-world scenes from video streams. To make immersive 3D video reality, the whole pipeline spanning from scene acquisition over 3D video reconstruction to real-time rendering needs to be researched. In this paper, we describe latest advancements of our system to record, reconstruct and render free-viewpoint videos of human actors. We apply a silhouette-based non-intrusive motion capture algorithm making use of a 3D human body model to estimate the actor’s parameters of motion from multi-view video streams. A renderer plays back the acquired motion sequence in real-time from any arbitrary perspective. Photo-realistic physical appearance of the moving actor is obtained by generating time-varying multi-view textures from video. This work shows how the motion capture sub-system can be enhanced by incorporating texture information from the input video streams into the tracking process. 3D motion fields are reconstructed from optical flow that are used in combination with silhouette matching to estimate pose parameters. We demonstrate that a high visual quality can be achieved with the proposed approach and validate the enhancements caused by the the motion field step.     

Video super-resolution reconstruction method based on deep Back projection and motion feature fusion

How to effectively utilize inter-frame redundancies is the key to improve the accuracy and speed of video super-resolution reconstruction methods. Previous methods usually process every frame in the whole video in the same way, and do not make full use of redundant information between frames, resulting in low accuracy or long reconstruction time. In this paper, we propose the idea of reconstructing key frames and non-key frames respectively, and give a video super-resolution reconstruction method based on deep back projection and motion feature fusion. Key-frame reconstruction subnet can obtain key frame features and reconstruction results with high accuracy. For non-key frames, key frame features can be reused by fusing them and motion features, so as to obtain accurate non-key frame features and reconstruction results quickly. Experiments on several public datasets show that the proposed method performs better than the state-of-the-art methods, and has good robustness.

     

Evaluation of an appearance-based 3D face tracker using dense 3D data

The ability to detect and track human heads and faces in video sequences can be considered as the finest level of any video surveillance system. In this paper, we introduce a general framework for evaluating our recent appearance-based 3D face tracker using dense 3D data. This tracker combines online appearance models with an image registration technique and can run in real-time and is drift insensitive. More precisely, accuracy and usability of this developed tracker are assessed using stereo-based range facial data from which ground truth 3D motions are computed. This evaluation quantifies the monocular tracker accuracy, and identifies its working range in 3D space. Additionally, this evaluation gives some hints on how the tracker can be fully exploited.     

基于特征分块的三维人脸重建和识别

针对传统三维人脸重建算法效率低且难以满足实际应用的缺陷,提出一种基于特征分块的三维人脸重建算法,并将此算法应用到三维人脸识别中,实现了基于特征分块的加权三维人脸识别。首先,利用基于平面模板的非均匀重采样法对原始数据进行归一化;其次,采用主动形状模型(ASM)算法对三维人脸和二维人脸图像进行特征定位和特征分块;然后,利用基于分块主元分析(PCA)的稀疏形变模型算法实现每个人脸分块的三维重建;最后,实现了此算法在三维人脸识别中的应用。实验表明,此重建算法具有较高的精度和重建效率,还可以达到全局最优,并且可以提高三维人脸的识别率。     

多模式3维视频形状编码

目的 具有立体感和高端真实感的3D视频正越来越受到学术界和产业界的关注和重视,未来在3D影视、机器视觉、远程医疗、军事航天等领域将有着广泛的应用前景。对象基3D视频是未来3D视频技术的重要发展趋势,其中高效形状编码是对象基3D视频应用中的关键问题。但现有形状编码方法主要针对图像和视频对象,面向3D视频的形状编码算法还很少。为此,基于对象基3D视频的应用需求,提出一种基于轮廓和链码表示的高效多模式3D视频形状编码方法。方法 对于给定的3D视频形状序列逐帧进行对象轮廓提取并预处理后,进行对象轮廓活动性分析,将形状图像分成帧内模式编码图像和帧间预测模式编码图像。对于帧内编码图像,基于轮廓内链码方向约束和线性特征进行高效编码。对于帧间编码图像,采用基于链码表示的轮廓基运动补偿预测、视差补偿预测、联合运动与视差补偿预测等多种模式进行编码,以充分利用视点内对象轮廓的帧间时域相关性和视点间对象轮廓的空域相关性,从而达到高效编码的目的。结果 实验仿真结果显示所提算法性能优于经典和现有的最新同类方法,压缩效率平均能提高9.3%到64.8%不等。结论 提出的多模式3D视频形状编码方法可以有效去除对象轮廓的帧间和视点间冗余,能够进行高效编码压缩,性能优于现有同类方法,可广泛应用于对象基编码、对象基检索、对象基内容分析与理解等。     

Examplar coherent 3D face reconstruction from forensic mugshot database

Reconstructing 3D face models from 2D face images is usually done by using a single reference 3D face model or some gender/ethnicity specific 3D face models. However, different persons, even those of the same gender or ethnicity, usually have significantly different faces in terms of their overall appearance, which forms the base of person recognition via faces. Consequently, existing 3D reference model based methods have limited capability of reconstructing precise 3D face models for a large variety of persons. In this paper, we propose to explore a reservoir of diverse reference models for 3D face reconstruction from forensic mugshot face images, where facial examplars coherent with the input determine the final shape estimation. Specifically, our 3D face reconstruction is formulated as an energy minimization problem with: 1) shading constraint from multiple input face images, 2) distortion and self-occlusion based color consistency between different views, and 3) depth uncertainty based smoothness constraint on adjacent pixels. The proposed energy is minimized in a coarse to fine way, where the shape refinement step is done by using a multi-label segmentation algorithm. Experimental results on challenging datasets demonstrate that the proposed algorithm is capable of recovering high quality 3D face models. We also show that our reconstructed models successfully boost face recognition accuracy.     

Constructing 3D facial hierarchical structure based on surface measurements

In this paper, we propose a novel framework for 3D facial similarity measures and facial data organization. The 3D facial similarity measures of our method are based on iso-geodesic stripes and conformal parameterization. Using the conformal parameterization, the 3D facial surface can be mapped into a 2D domain and the iso-geodesic stripes of the face can be measured. The measure results can be regarded as the similarity of faces, which is robust to head poses and facial expressions. Based on the measure result, a hierarchical structure of faces can be constructed, which is used to organize different faces. The structure can be utilized to accelerate the face searching speed in a large database. In experiment, we construct the hierarchical structures from two public facial databases: Gavab and Texas3D. The searching speed based on the structure can be increased by 4-6 times without accuracy loss of recognition.

     

Real-time 2D to 3D video conversion

We present a real-time implementation of 2D to 3D video conversion using compressed video. In our method, compressed 2D video is analyzed by extracting motion vectors. Using the motion vector maps, depth maps are built for each frame and the frames are segmented to provide object-wise depth ordering. These data are then used to synthesize stereo pairs. 3D video synthesized in this fashion can be viewed using any stereoscopic display. In our implementation, anaglyph projection was selected as the 3D visualization method, because it is mostly suited to standard displays.

DirectShow在三维视频纹理中的应用

首先介绍了DirectShow流媒体播放技术,在此基础上建立了一个自定义的渲染过滤器实现视频流数据的获取,并将此功能封装在动态链接库中,然后在实例程序中使用OpenGL将获取的视频流数据转换为实时的视频纹理,并显示在三维空间中呈二次曲面的物体的表面,实现了立体的、实时的和可控制的三维视频播放效果.该实例主要使用了DirectShow流媒体播放和智能连接技术、矩阵变换技术、三维纹理、二次曲面生成等技术.文章对于多媒体技术和三维仿真技术的综合应用具有一定的借鉴意义.     

一种自动鲁棒的三维人脸重建方法

三维人脸模型已经广泛应用到视频电话、视频会议、影视制作、电脑游戏、人脸识别等多个领域。目前三维人脸建模一般使用多幅图像，且要求表情中性。本文提出了基于正、侧面任意表情三维人脸重建方法。首先对二维图像中的人脸进行特征提取，然后基于三维人脸统计模型，通过缩放、平移、旋转等方法，及全局和局部匹配，获得特定的三维人脸。基于二维图像中的人脸纹理信息，通过纹理映射，获得完整的三维人脸。通过对大量实际二维人脸图像的三维人脸重建，证实了该方法的有效性和鲁棒性。     

单目视频中人体三维运动的迭代优化估计

提出并实现了一种从单目视频流中重建人体三维运动的方法．该方法通过交互定制得到个性化的人体骨架模型和视频序列每一帧中人体各关节点的二维坐标后,分别针对单帧和连续多帧进行优化并迭代求解,得到每一帧的比例因子的最优解;最后反求各关节点的三维坐标,重建人体三维运动序列．对包含复杂和快速多变的人体运动的视频进行的实验表明,该方法简单有效,适用于包括体育、影视等在内的实际视频源．     

基于弱监督学习的三维人脸形状与纹理重建

三维人脸相较于二维人脸包含了更多特征信息, 可应用于如人脸识别、影视娱乐、医疗美容等更多实际应用场景, 因此三维人脸重建技术一直是计算机视觉领域的研究热点. 由于真实三维人脸数据较难获取, 很多基于深度学习的重建算法首先利用传统重建方法为大量二维人脸图像构建三维标签, 作为训练数据, 这些数据可能并不精准, 从而导致算法的重建精度受到影响. 为此, 本文提出一种基于multi-level损失函数的弱监督学习模型, 结合传统三维人脸形变模型3DMM与深度学习方法, 直接从大量无三维标签的二维人脸图像中学习三维人脸特征信息, 从而实现基于单张二维人脸图像的三维人脸重建算法. 此外, 为解决二维人脸图像中常存在遮挡或大姿态情况而影响人脸纹理重建的问题, 本文使用基于CelebAMask-HQ数据集的人脸解析分割算法对图像进行预处理去除遮挡区域. 实验结果表明, 基于本文方法的三维人脸重建质量与重建精度均实现了一定的提升.     

Two Accelerating Techniques for 3D Reconstruction

Automatic reconstruction of 3D objects from 2D orthographic views has been a major research issue in CAD/CAM.In this paper,two acceleratin techniques to improve the efficiency of reconstruction are presented.First,some peudo elements are removed by depth and topology information as soon as the wire-frame is constructed ,which reduces the searching space.Second.the proposed algorithm does not establish all possible surfaces in the process of generating 3D faces.The surfaces and edge loops are generated by using the relationship between the boundaries of 3D faces and their projections,This avoids the growth in combinational complexity of previous methods that have to check all possible pairs of 3D candidate edges.     

Efficient 3D reconstruction for face recognition

Face recognition with variant pose, illumination and expression (PIE) is a challenging problem. In this paper, we propose an analysis-by-synthesis framework for face recognition with variant PIE. First, an efficient two-dimensional (2D)-to-three-dimensional (3D) integrated face reconstruction approach is introduced to reconstruct a personalized 3D face model from a single frontal face image with neutral expression and normal illumination. Then, realistic virtual faces with different PIE are synthesized based on the personalized 3D face to characterize the face subspace. Finally, face recognition is conducted based on these representative virtual faces. Compared with other related work, this framework has following advantages: (1) only one single frontal face is required for face recognition, which avoids the burdensome enrollment work; (2) the synthesized face samples provide the capability to conduct recognition under difficult conditions like complex PIE; and (3) compared with other 3D reconstruction approaches, our proposed 2D-to-3D integrated face reconstruction approach is fully automatic and more efficient. The extensive experimental results show that the synthesized virtual faces significantly improve the accuracy of face recognition with changing PIE.     

基于多尺度时域3D卷积的视频超分辨率重建

视频超分辨率是一项很有实用价值的工作.针对超高清产业中高分辨率资源较为匮乏的问题,为了有效利用视频序列帧间丰富的时间相关性信息及空间信息,提出一种基于多尺度时域3D卷积的视频超分辨率重建算法.该算法将输入的低分辨率视频序列帧分别通过不同时间尺度的3D卷积进行时空特征提取,3D卷积能够同时对空间与时间建模,相较于2D卷积...     

基于图像运动的三维重建及虚拟化应用

提出了一种稳定、快速地获取摄像机视频运动图像的三维重建方法,并对该运动图像做适当的虚拟化处理以展示重建效果。采用基于尺度不变特征点匹配的摄像机标定进行三维重建。尺度不变特征对于视频图像中的特征具有优秀敏锐的匹配能力,极大地放宽了摄像机标定对于设备上的限制,拓宽了实时三维重建的适用范围。通过对系统的一系列优化,不但提升了三维重建的精度,减少了错误匹配对摄像机标定的影响,而且进一步提升了处理速度。通过在三维重建的基础之上进行虚拟化处理,展示了本系统的三维重建效果。实验结果表明,该系统适用范围广,处理速度较快,重建精度高,实现了基于视频运动图像的三维重建。