Robust affine structure matching for 3D object recognition

We consider model-based object localization based on local geometric feature matching between the model and the image data. The method is based on geometric constraint analysis, working in transformation space. We present a formal method which guarantees finding all feasible matchings in polynomial time. From there we develop more computationally feasible algorithms based on conservative approximations of the formal method. Additionally, our formalism relates object localization, affine model indexing, and structure from multiple views to one another     

抓取操作中手与虚拟物体的自然融合

由于现有硬件和检测条件的限制,抓取中手指与虚拟物体之间往往会有穿透和空隙现象出现,无遮挡时就会被用户察觉;如果立即改变虚拟物体的位置来消除穿透和空隙,会使虚拟物体产生明显跳跃,缺乏交互真实感。分析了运动时手指与虚拟物体的遮挡关系,运用三维图形学中的理论实现对不遮挡情况的判断;将其与灰色预测理论相结合,实现对手的移动预测;根据预测结果和最大速度差原则,自适应地调整虚拟物体的速度,使虚拟物体与手逐渐融合。实验结果表明,使用提出的方法,既可提高抓取的成功率,又能保证虚拟物体与手的自然融合。     

Tracking a hand in interaction with an object based on single depth images

Multimedia Tools and Applications - Tracking a hand in interaction with an object based on vision is a challenging research topic. The occlusions that occur during the hand-object interaction make...     

M:N object matching between image and map object data sets by means of latent semantic analysis

In this article, we propose a method to find corresponding object-set pairs between image and map polygon object data sets by means of latent semantic analysis. Latent semantic analysis assigns each polygon object of both data sets to feature vectors in a continuous geometric space in which the similarities between the vectors are proportional to the priorities to constitute a corresponding object-set pair. Thus, object clusters can be obtained by applying an agglomerative hierarchical clustering to the feature vectors. These object clusters are separated into object-set pairs according to the data sets to which the objects belong and are evaluated with a geometric matching criterion to find corresponding object-set pairs. We applied the proposed method to the segmentation result of a composite image with six normalized difference vegetation index (NDVI) images and a forest inventory map. The proposed method was compared to a graph-embedding-based method. The results showed that the proposed method found more corresponding object-set pairs with a similar accuracy in terms of shape similarities and shared information of found pairs.     

Graph matching for object recognition and recovery

A robust skeleton-based graph matching method for object recognition and recovery applications is presented. The object model uses both a skeleton model and contour segment models, for object recognition and recovery. The presented skeleton-based shape matching method uses a combination of both structural and statistical methods that are applied in a sequential manner, which largely reduce the matching space when compared with previous works. This also provides a good alternate means to alleviate difficulties encountered in segmentation problems. Experiments of object recovery using real biomedical image samples have shown satisfactory results.     

Realtime and robust object matching with a large number of templates

Most of conventional object matching methods are based on comparing local features, which are too computational demanding. Recently, Dominant Orientation Templates (DOT) were proposed to solve the efficiency issue. Although DOT obtains promising results, it still suffers the problem of wasting too many bits in representation and fragility when partial occlusion occurs. As the number of templates increase, the performance will decrease. Therefore, we propose a compact DOT representation with a fast partial occlusion handling approach. Instead of using seven orientations in the original implementation, we employ single orientation of the highest gradients for the proposed compact DOT representation (C-DOT). Consequently, the size of feature vectors is reduced from 8 bits to 3 bits. To efficiently tackle the partial occlusion, we introduce the C-DOT similarity map to store the matching scores of individual grids in each sliding window, which is used to further infer the occlusion map. The experimental results demonstrate that the proposed method outperforms DOT.     

Ranking consistency for image matching and object retrieval

The goal of object retrieval is to rank a set of images by the similarity of their contents to those of a query image. However, it is difficult to measure image content similarity due to visual changes caused by varying viewpoint and environment. In this paper, we propose a simple, efficient method to more effectively measure content similarity from image measurements. Our method is based on the ranking information available from existing retrieval systems. We observe that images within the set which, when used as queries, yield similar ranking lists are likely to be relevant to each other and vice versa. In our method, ranking consistency is used as a verification method to efficiently refine an existing ranking list, in much the same fashion that spatial verification is employed. The efficiency of our method is achieved by a list-wise min-Hash scheme, which allows rapid calculation of an approximate similarity ranking. Experimental results demonstrate the effectiveness of the proposed framework and its applications.     

Fast and accurate extraction of moving object silhouette for personalized Virtual Reality Studio @ Home

Accurate segmentation of moving object silhouette in a real-time video is very important for object silhouette extraction in the vision-based interactive systems. However, the inherent problem of moving object segmentation based on the background subtraction criteria is to distinguish the changes occurring from background disturbing effects such as noise, shadows and illumination changes. The present paper proposes a hybrid method based on the background subtraction criteria that preserves the boundary of moving object and also robust against the noise and illumination changes. In the proposed method, the object regions are well identified by fusing the results from the background difference and motion-based change detection criterion. The shadows and highlights are well detected by utilizing the normalized luminance and background difference in Hue and Saturation component. The paper also introduces a novel connected component analysis procedure for detecting the object blob from the noise blobs, and a robust pixel-based background update scheme for updating the dynamic changes in the background. Moreover, the computational complexity of the proposed algorithm is analyzed. The proposed method has been implemented and evaluated regarding the segmentation quality and the frame rate. Further, the method has been shown to successfully extract the moving object silhouette and robust against the disturbing effects. Moreover, the proposed method has been tested in the VR@Home platform.     

Analysis of a manipulator in relation to collision between a link and an object

In this article, equations of motion of a manipulator are derived after consideration of the characteristics of the driving source. By considering a collision between a link and an object, and considering the active motion to absorb the kinetic energy of the object, the trajectories for saving energy are calculated by the iterative dynamic programming (IDP) method. The dynamic characteristics of manipulator control based on the trajectory for saving energy are also analyzed theoretically and investigated experimentally.     

Shapeme histogram projection and matching for partial object recognition

Histograms of shape signature or prototypical shapes, called shapemes, have been used effectively in previous work for 2D/3D shape matching and recognition. We extend the idea of shapeme histogram to recognize partially observed query objects from a database of complete model objects. We propose representing each model object as a collection of shapeme histograms and match the query histogram to this representation in two steps: 1) compute a constrained projection of the query histogram onto the subspace spanned by all the shapeme histograms of the model and 2) compute a match measure between the query histogram and the projection. The first step is formulated as a constrained optimization problem that is solved by a sampling algorithm. The second step is formulated under a Bayesian framework, where an implicit feature selection process is conducted to improve the discrimination capability of shapeme histograms. Results of matching partially viewed range objects with a 243 model database demonstrate better performance than the original shapeme histogram matching algorithm and other approaches.     

Investigation of hand postures in manufacturing industries according to hand and object properties

The aim of this study was to investigate the hand postures commonly used in manufacturing companies, according to the use of either the right or left hand and the object properties. Operations taking place over 1636 s were reviewed to record hand postures, right and left hands, object shapes, object locations, and object directions and to classify them according to a hand posture classification system developed in this study. The classification system considers R (resting), G (grasping), P (pinching), and T (touching), which are further subdivided based on the number and use of fingers-(T) thumb, (I) index, (M) middle, (R) ring, or (L) little-and palm (P). The hand postures classified as 5G (TIMRL), 3P (TIM), 5P (TIMRL), 2P (TI), and 4G (TMRL) were the most commonly used postures; however, their frequencies of use depended on whether the right or left hand was used and the object properties. The most commonly handled object shapes were cylinders, rectangles, and sheets, and the most common location for objects of these shapes was on a table. The use of hand postures depends on the type of activity. Unlike activities of daily living, which often require pinching an object, grasping is more often used in the manufacturing industry to manipulate hand tools and parts with force. The results of this study are anticipated to be useful in designing future research studies on hand postures.Relevance to industryThe investigation of commonly used hand postures is of special interest in ergonomics because of its association with musculoskeletal disorders of the hand. Information on commonly used hand posture can be used in designing future hand posture research studies to estimate and reduce hand stresses.     

Online construction of the manipulation model of an unknown object by a robot hand using the regrasping primitives

This paper presents a planner which explores an unknown polyhedral object with a multifingered hand in a systematic way by constructing the manipulation-model of the object. The manipulation-model of an object we proposed previously is a database which consists of landmark orientations of the object, the applicability of manipulation primitives to them and intertransformations among them. In this paper, two regrasping primitives, rotation and pivoting, are employed. A multifingered hand probes a given unknown object for pivot/rotation axes. If a new axis is found, the planner recursively performs pivoting or rotation-test (which is similar to rotation but does not actually rotate the object) about the found axis and searches the object for more axes. The manipulation-model of the object is constructed based on the found axes. An experiment for constructing a manipulation-model of a non-convex polyhedral object is conducted using a fourfingered hand. Force/torque sensors mounted on the fingertips collect the data of the object. A geometrical model of the object is also constructed.     

Object recognition and articulated object learning by accumulative Hopfield matching

In this paper, a novel object recognition method based on attributed relational graph matching is proposed, which is called accumulative Hopfield matching. We first divide the scene graph into many sub-graphs, and a modified Hopfield network is then constructed to obtain the sub-graph isomorphism between each sub-scene graph and model graph. The final result is deduced by accumulating the solutions of all small sub-networks. Comparing to the traditional Hopfield network, the proposed system has the advantage of finding homomorphic mappings between two graphs. Furthermore, the system can be applied for articulated object recognition and visual model learning, which is considered as a difficult topic till now. The proposed method has been evaluated with real images.     

Analysis of a manipulator in consideration of the relative motion between a tray and an object

In this article, equations of motion of a manipulator, whose mechanism has a tray installed with a passive revolute joint, are derived after consideration of the characteristics of the driving source. Considering the relative motion between the tray and the object, the trajectories of the velocity for saving energy are calculated by iterative dynamic programming. Also, the dynamic characteristics of manipulator control based on the trajectory for saving energy are analyzed theoretically and investigated experimentally.     

Three-dimensional finger joint angles by hand posture and object properties

The objective of this study was to identify three-dimensional finger joint angles for various hand postures and object properties. Finger joint angles were measured using a VICON system for 10 participants while they pinched objects with two, three, four and five fingers and grasped them with five fingers. The objects were cylinders and square pillars with diameters of 2, 4, 6 and 8 cm and weights of 400, 800, 1400 and 1800 g. Hand posture and object size more significantly affected the joint flexion angles than did object shape and weight. Object shape affected only the metacarpophalangeal (MCP) joint angle of the index finger and the flexion angle of the MCP joint of the little finger. Larger flexion angles resulted when the hand posture was grasping with five fingers. The joint angle increased linearly as the object size decreased. This report provides fundamental information about the specific joint angles of the thumb and fingers.

Practitioner Summary: Three-dimensional finger joint angles are of special interest in ergonomics because of their importance in handheld devices and musculoskeletal hand disorders. In this study, the finger joint angles corresponding to various hand postures and objects with different properties were determined.     

Reeb graph path dissimilarity for 3D object matching and retrieval

We introduce a skeletal graph for topological 3D shape representation using Morse theory. The proposed skeletonization algorithm encodes a 3D shape into a topological Reeb graph using a normalized mixture distance function. We also propose a novel graph matching algorithm by comparing the relative shortest paths between the skeleton endpoints. Experimental results demonstrate the feasibility of the proposed topological Reeb graph as a shape signature for 3D object matching and retrieval.     

基于SIFT特征匹配与K-均值聚类的运动目标检测

运动摄像机情况下的运动目标检测是视频监控中的难点和热点问题。为了能够有效地检测出运动目标,根据视频中背景与运动目标的速度不同这一特点,提出了一个基于尺寸不变特征变换(SIFT)和K-均值聚类的运动目标检测方法。首先提取视频中相邻两帧图像的SIFT特征点并进行匹配,并计算匹配特征点的运动速度,最后将运动目标和背景上的SIFT特征点K-均值聚类分析,在单运动目标、多运动目标和带有摄像头旋转情况下做了实验。实验结果表明,提出的目标检测算法能够在运动背景下较好地检测到目标并保留稳定的目标局部特征,对于摄像机运动、摄像机旋转、亮度变化等影响因素具有较强的适应能力。