一种基于极坐标变换的点模式匹配算法

点模式匹配是目标识别、图像配准与匹配、姿态估计等计算机视觉与模式识别应用方向的基础问题之一。提出了一种新的利用点特征进行匹配的算法,该算法根据点集的分布与点位置信息,构建了点的特征属性图,通过极坐标变换得到对数极坐标的特征图,并利用几何不变矩方法对特征图进行描述。由特征描述向量的比较,获得粗匹配结果,然后通过几何约束迭代的方法获取最终的点集匹配结果。本文贡献如下:一,构建了一种点的极坐标变换特征,并运用不变矩进行描述,使所提特征具有旋转与平移的不变性;二,提出了利用点特征与整体点集几何约束结合的匹配算法,能有效克服出格点与噪声带来的不利影响。最终实验说明了算法的有效性和鲁棒性。     

基于移动散射点模型的雷达回波仿真及分析

针对中段目标宽带雷达回波信号难于获取的问题,进一步研究了目标雷达视线角和中段目标姿态的建模方法,给出了基于移动散射点模型的散射中心位置计算公式,然后基于几何绕射理论计算了各散射中心的散射强度,最后对雷达一维距离像进行归一化处理,提取了散射点的相对距离特征,得到了散射点相对距离变化的特征序列.实验结果显示,对雷达一维距离像进行特征提取,可以得到中段目标的姿态信息,为中段目标识别奠定基础.     

相似物体的姿态矫正算法设计及实现

在三维点云识别过程中,由于点云姿态不同,比较点的特征时,要将空间姿态变换计算在内,旧的空间姿态算法复杂度高,计算量大。针对上述问题,提出一种相似物体的姿态矫正算法,通过在两个相似物体要进行识别的位置上选择不共线的三组对应点,经过空间变换与旋转,将相似部位的姿态统一,为物体识别做基础。通过使用汽车点云模型进行矫正,验证了该算法的可行性和正确性。     

信号博弈的三维重建形变识别算法

现有双目立体视觉算法常常需要双目相机位置固定,在现实应用中,这类算法难以重构空间三维几何关系.为此,本文提出了一种不受位置限制的多视角图像三维重建与形变检测算法.该算法首先采用sift算法获取成对图像的特征点,以获取形变前后比对点对图像的特征信息;其次,利用信号博弈方法确定图像拍摄时相机的空间位置与视角,以准确获取图像的空间位置坐标;再次,依据上述信息完成物体的三维点云重建;进而,利用三维数据信息比对实现物体形变识别.最后,本文利用真实物体的实验,验证了三维重建形变识别算法的有效性.     

基于几何证据收集的物体检测

人造物体通常具有非常稳定的形状特征,这种几何上较为持久和稳定的属性为物体识别提供了证据,并且它们相对于外观特征、颜色特征或灰度梯度特征等更具有稳定性和分辨力。基于形状特征对物体进行识别的难点在于物体的颜色、光照、尺寸、位置、姿态和背景干扰总是在不断变化的,且不可能事先预测到所有可能的环境。这种物体本身和环境的多样性使得基于几何形状的物体识别成为一个非常具有挑战性的难题。通过定义一种基于形状模板、对边缘线段化之后的图像进行几何证据筛选、收集和组合判断的方法,实现从背景环境中精确找到目标物体,并能够指出组成物体的线段的语义属性。该方法的实质是解一个全局最优的组合优化问题。虽然全局最优组合优化问题看似复杂度很高,但它无需定义复杂的特征向量,无需高代价的样本训练过程,具有非常好的泛化能力和环境适应性,并且能够使用几何特征来提高组合效率,具有更可靠的认知心理学依据。此外,几何证据收集过程简洁明了及具有普适性的特点使其表现出极大的应用前景。实验结果证明该方法在应对环境变化、不变性识别、精确指出物体的几何构造、搜索效率与计算量等方面表现出显著的优势。这一尝试有助于理清发生在物体识别过程中的一些带有普遍性的加工环节。     

基于人造物体直线段结构特征的不变性识别

传统的用假设验证法进行三维物体识别的方法需要通过一组非线性方程组求解从模型到场景的坐标系变换,具有非常高的复杂度.文中提出了一种基于能够表明物体几何构造的直线段特征的人造物体识别方法,将假设验证法中对于全局坐标系变换的求解分散在各个平面单应性变换的求解中,降低了求解的复杂度.该方法首先利用几何不变量预匹配特征点,进而假设并求出场景和模型平面之间的单应矩阵,随后通过模型与场景之间直线段特征匹配的结果进行验证.实验证明,该方法能够快速准确地识别含有较多共面直线段特征的人造物体.     

基于多重几何约束的未知物体抓取位姿估计

针对机器人在非结构化环境下面临的未知物体难以快速稳定抓取的问题,提出一种基于多重几何约束的未知物体抓取位姿估计方法.通过深度相机获取场景的几何点云信息,对点云进行预处理得到目标物体,利用简化的夹持器几何形状约束生成抓取位姿样本.然后,利用简化的力封闭约束对样本进行快速粗筛选.对抓取位姿的抓取几何轮廓进行力平衡约束分析,将稳定的位姿传送至机器人执行抓取.采用深度相机与6自由度机械臂组成实验平台,对不同姿态形状的物体进行抓取实验.实验结果表明,本文方法能够有效应对物体种类繁多、缺乏3维模型的情况,在单目标和多目标场景均具有良好的适用性.     

基于多相机的人脸姿态识别

主动形状模型(ASM)算法被用来进行人脸特征点的精确定位,然后在多相机测量的图像中进行特征点的立体匹配,利用双目视觉和相机三维测距技术可以确定人脸特征点的空间三维位置,从而利用这些特征点的相对位置确定出人脸的姿态。实验结果显示,用该方法进行人脸姿态识别能取得比二维识别更高的精确度。     

基于焊件识别与位姿估计的焊接机器人视觉引导

为实现智能化柔性化焊接,并解决焊件本身立体结构遮挡焊缝影响视觉引导的问题,提出一种基于物体识别与位姿估计思想的焊接机器人视觉引导方法.首先,通过焊件的CAD模型离线建立焊接模型库;然后,在线计算焊件点云的VFH特征,与焊接模型库进行比对实现焊件识别,计算FPFH特征实现SAC-IA和ICP结合的两步位姿估计,并通过假设验证优化结果;最后,利用焊接模型库的焊接信息结合焊件位姿生成焊接轨迹,为焊接机器人提供视觉引导.实验结果表明,所提方法可以利用目标整体三维信息准确识别焊件并估计其姿态,进而引导机器人完成智能柔性的焊接操作.     

区域高斯描绘子及其在物体识别中的应用

为了对几何变形的图像进行正确和有效的识别,对基于物体轮廓的高斯描绘子进行推广,构造了一种基于区域的新的不变量--区域高斯描绘子.构造思路主要有两点:①定义一个具有平移、尺度不变性的函数--区域高斯势函数;②分别计算区域高斯势函数在8个同心圆上的平均值,从而获得8个不变量.这些不变量不仅具有平移、旋转、尺度和反射不变性,而且对噪声不敏感,适用范围广.将区域高斯描绘子应用于物体识别,获得很高的识别率.     

Car model recognition by utilizing symmetric property to overcome severe pose variation

This paper presents a mirror morphing scheme to deal with the challenging pose variation problem in car model recognition. Conventionally, researchers adopt pose estimation techniques to overcome the pose problem, whereas it is difficult to obtain very accurate pose estimation. Moreover, slight deviation in pose estimation degrades the recognition performance dramatically. The mirror morphing technique utilizes the symmetric property of cars to normalize car images of any orientation into a typical view. Therefore, the pose error and center bias can be eliminated and satisfactory recognition performance can be obtained. To support mirror morphing, active shape model (ASM) is used to acquire car shape information. An effective pose and center estimation approach is also proposed to provide a good initialization for ASM. In experiments, our proposed car model recognition system can achieve very high recognition rate (>95%) with very low probability of false alarm even when it is dealing with the severe pose problem in the cases of cars with similar shape and color.     

三维人体姿态估计研究综述

三维人体姿态估计在本质上是一个分类问题和回归问题,主要通过图像估计人体的三维姿态。基于传统方法和深度学习方法的三维人体姿态估计是当前研究的主流方法。按照传统方法到深度学习方法的顺序对近年来三维人体姿态估计方法进行系统介绍,从而了解传统方法通过生成和判别等方法得到人体姿态的众多要素完成三维人体姿态的估计。基于深度学习的三维人体姿态估计方法主要通过构建神经网络,从图像特征中回归出人体姿态信息,大致可以分为基于直接回归方法、基于2D信息方法和基于混合方法的三维人体姿态估计这三类。最后对当前三维人体姿态估计研究所面临的困难与挑战进行阐述,并对未来的研究趋势做出展望。     

Trajectory reconstruction with uncertainty estimation using mosaic registration

This paper addresses the problem of estimating the 3D trajectory and associated uncertainty of an underwater autonomous vehicle from a set of images of the seabed taken by an onboard camera. The presented algorithms resort to the use of video mosaics and build upon previous work on image registration and visual pose estimation. The pose estimation is accomplished in two steps. Firstly, a video mosaic is created automatically, covering a region of interest of the seabed. Then, after associating a 3D referential for the mosaic, the estimation of the camera position from a new view of the scene becomes possible.

The main contribution of this paper lies on the assessment of the performance of the 3D pose algorithms. In order to do this, an image sequence with available ground-truth is used for precise error measuring. A first-order error propagation analysis is presented, relating the uncertainty in the location of the match points with the uncertainty in the pose parameters. The importance of predicting the estimate uncertainty is emphasized by the fact that it can be used for comparing algorithms and for the on-line monitoring of the vehicle trajectory reconstruction quality.

Several iterative and non-iterative pose estimation methods are discussed, differing both on the criteria being minimized and on the required information about the camera intrinsic parameters. This information ranges from the full knowledge of the parameters, to the case where they are estimated using self-calibration from an image sequence under pure rotation. The implemented pose algorithms are compared for the accuracy and estimate covariance.     

结合稀疏表示和深度学习的视频中3D人体姿态估计

目的 2D姿态估计的误差是导致3D人体姿态估计产生误差的主要原因,如何在2D误差或噪声干扰下从2D姿态映射到最优、最合理的3D姿态,是提高3D人体姿态估计的关键。本文提出了一种稀疏表示与深度模型联合的3D姿态估计方法,以将3D姿态空间几何先验与时间信息相结合,达到提高3D姿态估计精度的目的。方法 利用融合稀疏表示的3D可变形状模型得到单帧图像可靠的3D初始值。构建多通道长短时记忆MLSTM（multi-channel long short term memory）降噪编/解码器,将获得的单帧3D初始值以时间序列形式输入到其中,利用MLSTM降噪编/解码器学习相邻帧之间人物姿态的时间依赖关系,并施加时间平滑约束,得到最终优化的3D姿态。结果 在Human3.6M数据集上进行了对比实验。对于两种输入数据：数据集给出的2D坐标和通过卷积神经网络获得的2D估计坐标,相比于单帧估计,通过MLSTM降噪编/解码器优化后的视频序列平均重构误差分别下降了12.6%,13%;相比于现有的基于视频的稀疏模型方法,本文方法对视频的平均重构误差下降了6.4%,9.1%。对于2D估计坐标数据,相比于现有的深度模型方法,本文方法对视频的平均重构误差下降了12.8%。结论 本文提出的基于时间信息的MLSTM降噪编/解码器与稀疏模型相结合,有效利用了3D姿态先验知识,视频帧间人物姿态连续变化的时间和空间依赖性,一定程度上提高了单目视频3D姿态估计的精度。     

Instance segmentation based 6D pose estimation of industrial objects using point clouds for robotic bin-picking

3D object pose estimation for robotic grasping and manipulation is a crucial task in the manufacturing industry. In cluttered and occluded scenes, the 6D pose estimation of the low-textured or textureless industrial object is a challenging problem due to the lack of color information. Thus, point cloud that is hardly affected by the lighting conditions is gaining popularity as an alternative solution for pose estimation. This article proposes a deep learning-based pose estimation using point cloud as input, which consists of instance segmentation and instance point cloud pose estimation. The instance segmentation divides the scene point cloud into multiple instance point clouds, and each instance point cloud pose is accurately predicted by fusing the depth and normal feature maps. In order to reduce the time consumption of the dataset acquisition and annotation, a physically-simulated engine is constructed to generate the synthetic dataset. Finally, several experiments are conducted on the public, synthetic and real datasets to verify the effectiveness of the pose estimation network. The experimental results show that the point cloud based pose estimation network can effectively and robustly predict the poses of objects in cluttered and occluded scenes.     

Pose estimation using line-based dynamic vision and inertial sensors

An observer problem from a computer vision application is studied. Rigid body pose estimation using inertial sensors and a monocular camera is considered and it is shown how rotation estimation can be decoupled from position estimation. Orientation estimation is formulated as an observer problem with implicit output where the states evolve on SO(3). A careful observability study reveals interesting group theoretic structures tied to the underlying system structure. A locally convergent observer where the states evolve on SO (3) is proposed and numerical estimates of the domain of attraction is given. Further, it is shown that, given convergent orientation estimates, position estimation can be formulated as a linear implicit output problem. From an applications perspective, it is outlined how delayed low bandwidth visual observations and high bandwidth rate gyro measurements can provide high bandwidth estimates. This is consistent with real-time constraints due to the complementary characteristics of the sensors which are fused in a multirate way.     

Exact and approximate solutions of the perspective-three-pointproblem

Model-based pose estimation techniques that match image and model triangles require large numbers of matching operations in real-world applications. The authors show that by using approximations to perspective, 2D lookup tables can be built for each of the triangles of the models. An approximation called `weak perspective' has been applied previously to this problem; the authors consider two other perspective approximations: paraperspective and orthoperspective. These approximations produce lower errors for off-center image features than weak perspective     

Human steering angle estimation in video based on key point detection and Kalman filter

Human pose recognition and estimation in video is pervasive. However, the process noise and local occlusion bring great challenge to pose recognition. In this paper, we introduce the Kalman filter into pose recognition to reduce noise and solve local occlusion problem. The core of pose recognition in video is the fast detection of key points and the calculation of human steering angles. Thus, we first build a human key point detection model. Frame skipping is performed based on the Hamming distance of the hash value of every two adjacent frames in video. Noise reduction is performed on key point coordinates with the Kalman filter. To calculate the human steering angle, current state information of key points is predicted using the optimal estimation of key points at the previous time. Then human steering angle can be calculated based on current and previous state information. The improved SENet, NLNet and GCNet modules are integrated into key point detection model for improving accuracy. Tests are also given to illustrate the effectiveness of the proposed algorithm.     

Pose Estimation using Point and Line Correspondences

The problem of a real-time pose estimation between a 3D scene and a single camera is a fundamental task in most 3D computer vision and robotics applications such as object tracking, visual servoing, and virtual reality. In this paper we present two fast methods for estimating the 3D pose using 2D to 3D point and line correspondences. The first method is based on the iterative use of a weak perspective camera model and forms a generalization of DeMenthon's method (1995) which consists of determining the pose from point correspondences. In this method the pose is iteratively improved with a weak perspective camera model and at convergence the computed pose corresponds to the perspective camera model. The second method is based on the iterative use of a paraperspective camera model which is a first order approximation of perspective. We describe in detail these two methods for both non-planar and planar objects. Experiments involving synthetic data as well as real range data indicate the feasibility and robustness of these two methods. We analyse the convergence of these methods and we conclude that the iterative paraperspective method has better convergence properties than the iterative weak perspective method. We also introduce a non-linear optimization method for solving the pose problem.