Recurrent Motion Refiner for Locomotion Stitching

Stitching different character motions is one of the most commonly used techniques as it allows the user to make new animations that fit one's purpose from pieces of motion. However, current motion stitching methods often produce unnatural motion with foot sliding artefacts, depending on the performance of the interpolation. In this paper, we propose a novel motion stitching technique based on a recurrent motion refiner (RMR) that connects discontinuous locomotions into a single natural locomotion. Our model receives different locomotions as input, in which the root of the last pose of the previous motion and that of the first pose of the next motion are aligned. During runtime, the model slides through the sequence, editing frames window by window to output a smoothly connected animation. Our model consists of a two-layer recurrent network that comes between a simple encoder and decoder. To train this network, we created a sufficient number of paired data with a newly designed data generation. This process employs a K-nearest neighbour search that explores a predefined motion database to create the corresponding input to the ground truth. Once trained, the suggested model can connect various lengths of locomotion sequences into a single natural locomotion.     

基于四元数描述和EMD的人体运动捕获数据检索

针对运动捕获数据的高效匹配问题,提出了一种新的基于四元数描述和EMD（ Earth Mover＆#39;s Distance）的人体运动检索算法。该算法主要包括特征提取和运动匹配两部分。在特征提取部分,为了解决高维数据检索效率低的问题,引入了四元数描述符对关节点的数据信息特征进行描述,通过映射姿态分布的原始数据,并采取K-means聚类方法对待查询动作和运动数据库的特征数据进行降维并归类。在运动匹配部分,根据聚类结果,建立每个特征数据集的距离矩阵,将匹配问题转换为运输优化问题。然后,用EMD算法度量待查询动作和数据库动作之间的相似值。仿真实验结果证明了提出的算法是有效的。     

Human motion retrieval using topic model

Content‐based human motion retrieval is important for animators with the development of motion editing and synthesis, which need to search similar motions in large databases. Obtaining text‐based representation from quantization of mocap data turned out to be efficient. It becomes a fundamental step of many researches in human motion analysis. Geometric features are one of these techniques, which involve much prior knowledge and reduce data redundancy of numerical data. We describe geometric features as basic unit to define human motions (also called mo‐words) and view a human motion as a generative process. Therefore, we obtain topic motions, which possess more semantic information using latent Dirichlet allocation by learning from massive training examples in order to understand motions better. We combine probabilistic model with human motion retrieval and come up with a new representation of human motions and a new retrieval framework. Our experiments demonstrate its advantages, both for understanding motions and retrieval. Copyright © 2011 John Wiley & Sons, Ltd.     

Human motion prediction for human-robot collaboration

In human-robot collaborative manufacturing, industrial robots would work alongside human workers who jointly perform the assigned tasks seamlessly. A human-robot collaborative manufacturing system is more customised and flexible than conventional manufacturing systems. In the area of assembly, a practical human-robot collaborative assembly system should be able to predict a human worker’s intention and assist human during assembly operations. In response to the requirement, this research proposes a new human-robot collaborative system design. The primary focus of the paper is to model product assembly tasks as a sequence of human motions. Existing human motion recognition techniques are applied to recognise the human motions. Hidden Markov model is used in the motion sequence to generate a motion transition probability matrix. Based on the result, human motion prediction becomes possible. The predicted human motions are evaluated and applied in task-level human-robot collaborative assembly.     

基于约束轨迹重构的运动编辑

提出了一种运动数据重用的方法.采用偏移映射技术构建模型末端的位姿约束轨迹,该轨迹较好地保留了原始运动的特征;利用Unscented卡尔曼滤波方法求解约束,可以实时地生成与原始运动相似的新运动.实验结果表明,该方法可以生成满足不同模型和场景要求的多种运动,扩展了捕获数据的应用范围.     

Let's be a virtual juggler

Juggling, which uses both hands to keep several objects in the air at once, is admired by anyone who sees it. However, skillful real‐world juggling requires long, hard practice. Therefore, we propose an interesting method to enable anyone to juggle skillfully in the virtual world. In the real world, the human motion has to follow the motion of the moving objects; in the virtual world, the objects' motion can be adjusted together with the human motion. By using this freedom, we have generated a juggling avatar that can follow the user's motion. The user simply makes juggling‐like motions in front of a motion sensor. Our system then searches for juggling motions that closely match the user's motions and connects them smoothly. We then generate moving objects that both satisfy the laws of physics and are synchronized with the synthesized motion of the avatar. In this way, we can generate a variety of juggling animations by an avatar in real time. Copyright © 2016 John Wiley & Sons, Ltd.     

Combining inverse blending and Jacobian‐based inverse kinematics to improve accuracy in human motion generation

We present in the paper a hybrid method for motion editing combining motion blending and Jacobian‐based inverse kinematics (IK). When the original constraints are changed, a blending‐based IK solver is first employed to find an adequate joint configuration coarsely. Using linear motion blending, this search corresponds to a gradient‐based minimization in the weight space. The found solution is then improved by a Jacobian‐based IK solver by further minimizing the distance between the end effectors and constraints. To accelerate the searching in the weight space, we introduce a weight map, which pre‐computes the good starting positions for the gradient‐based minimization. The advantages of our approach are threefold: first, more realistic motions can be generated by utilizing motion blending techniques, compared with pure Jacobian‐based IK. The blended results also increase the rate of convergence of the Jacobian‐based IK solver. Second, the Jacobian‐based IK solver modifies poses in the pose configuration space and the computational cost does not scale with the number of examples. Third, it is possible to extrapolate the given example motions with a Jacobian‐based IK solver, while it is generally difficult with pure blending‐based techniques. Copyright © 2014 John Wiley & Sons, Ltd.     

N-body spacetime constraints

Animators frequently choreograph complex motions for multiple objects that interact through collision and obstruction. In such situations, the use of physically based dynamics to confer visual realism creates challenging computational problems. Typically forward simulation is well understood, but the inverse problem of motion synthesis—that of synthesizing motions consistent both with physical law and with the animator's requirements—is generally tedious and sometimes intractable. We show how N-body inverse problems can be formulated as optimization tasks. We present a simply stated, but combinatorially formidable example that exhibits all of the essential sources of complexity common to N-body motion synthesis, and show how it can be solved approximately using heuristic methods based on evolutionary computation.     

Enriching a motion database by analogous combination of partial human motions

We have synthesized new human body motions from existing motion data, by dividing the body of an animated character into several parts, such as upper and lower body, and partitioning the motion of the character into corresponding partial motions. By combining different partial motions, we can generate new motion sequences. We select the most natural-looking combinations by analyzing the similarity of partial motions, using techniques such as motion segmentation, dimensionality reduction, and clustering. These new combinations can dramatically increase the size of a motion database, allowing more score in selecting motions to meet constraints, such as collision avoidance. We verify the naturalness and physical plausibility of the new motions using an SVM learning model and by analysis of static and dynamic balance.     

Computer animation of human walking: a survey

This paper surveys the set of techniques developed in computer graphics for animating human walking. First we focus on the evolution from purely kinematic ‘knowledge‐based’ methods to approaches that incorporate dynamic constraints or use dynamic simulations to generate motion. Then we review the recent advances in motion editing that enable the control of complex animations by interactively blending and tuning synthetic or captured motions. Copyright © 1999 John Wiley & Sons, Ltd.     

Finding Syntactic Structures from Human Motion Data

We present a new approach to motion rearrangement that preserves the syntactic structures of an input motion automatically by learning a context‐free grammar from the motion data. For grammatical analysis, we reduce an input motion into a string of terminal symbols by segmenting the motion into a series of subsequences, and then associating a group of similar subsequences with the same symbol. To obtain the most repetitive and precise set of terminals, we search for an optimial segmentation such that a large number of subsequences can be clustered into groups with little error. Once the input motion has been encoded as a string, a grammar induction algorithm is employed to build up a context‐free grammar so that the grammar can reconstruct the original string accurately as well as generate novel strings sharing their syntactic structures with the original string. Given any new strings from the learned grammar, it is straightforward to synthesize motion sequences by replacing each terminal symbol with its associated motion segment, and stitching every motion segment sequentially. We demonstrate the usefulness and flexibility of our approach by learning grammars from a large diversity of human motions, and reproducing their syntactic structures in new motion sequences.     

基于多自主智能体的群体动画创作

群体动画一直是计算机动画界一个具有挑战性的研究方向，提出了一个基于多自主智能体的群体动画创作框架：群体中的各角色作为自主智能体，能感知环境信息，产生意图，规划行为，最后通过运动系统产生运动来完成行为和实现意图，与传统的角色运动生成机理不同，首先采用运动捕获系统建立基本运动库，然后通过运动编辑技术对基本运动进行处理以最终得到角色运动，应用本技术，动画师只需“拍摄”角色群体的运动就能创作群体动画，极大地提高了制作效率。     

Acquisition of obstacle motion patterns to improve mobile robot motion planning

Mobile robots for advanced applications have to act in environments which contain moving obstacles (humans). Even though the motions of such obstacles are not precisely predictable, usually they are not completely random; long-term observation of obstacle behavior may thus yield valuable knowledge about prevailing motion patterns. By incorporating such knowledge as statistical data, a new approach called statistical motion planning yields robot motions which are better adapted to the dynamic environment. To put these ideas into practice, an experimental system has been developed. Cameras observe the workspace in order to detect obstacle motion. Statistical data is derived and represented as a set of stochastic trajectories. This data can be directly employed in order to calculate collision probability, i.e. the probability of encountering an obstacle during the robot's motion. Further aspects of motion planning are addressed: path planning which minimizes collision probability, estimation of expected time to reach the goal and reactive planning.     

Skiing Simulation Based on Skill‐Guided Motion Planning

Skiing is a popular recreational sport, and competitive skiing has been events at the Winter Olympic Games. Due to its wide moving range in the outdoor environment, motion capture of skiing is hard and usually not a good solution for generating skiing animations. Physical simulation offers a more viable alternative. However, skiing simulation is challenging as skiing involves many complicated motor skills and physics, such as balance keeping, movement coordination, articulated body dynamics and ski‐snow reaction. In particular, as no reference motions — usually from MOCAP data — are readily available for guiding the high‐level motor control, we need to synthesize plausible reference motions additionally. To solve this problem, sports techniques are exploited for reference motion planning. We propose a physics‐based framework that employs kinetic analyses of skiing techniques and the ski–snow contact model to generate realistic skiing motions. By simulating the inclination, angulation and weighting/unweighting techniques, stable and plausible carving turns and bump skiing animations can be generated. We evaluate our framework by demonstrating various skiing motions with different speeds, curvature radii and bump sizes. Our results show that employing the sports techniques used by athletes can provide considerable potential to generate agile sport motions without reference motions.     

Development of a muscle suit for the upper body—realization of abduction motion

A 'muscle suit' that will provide muscular support for the paralyzed or those otherwise unable to move unaided is being developed. The muscle suit is a garment without a metal frame that uses McKibben actuators driven by compressed air to produce motion. Because the actuators are sewn into the garment, no metal frame is needed, making the muscle suit very light and cheap. These features are completely different from conventional methods for supporting humans. In this paper, the basic concept and advantages of the muscle suit are first described. In order to verify the feasibility of the basic concept, a prototype system is mounted on a life-size doll. This allows for the limitations of the original design to be identified. Next, an armor-type muscle suit is proposed in order to overcome some of the prototype's limitations. A full-range abduction motion, which is determined to be the most difficult upper body motion, is realized through geometric analysis and experimentation. In the future, the remaining arm motions will be investigated and implemented.     

Estimate Large Motions Using the Reliability-Based Motion Estimation Algorithm

Detecting and estimating motions of fast moving objects has many important applications. However, most existing motion estimation techniques have difficulties in handling large motions in the scene. In this paper, we extend our recently proposed reliability-based stereo vision technique to solving large motion estimation problem. Compared with our stereo vision approach, the new algorithm removes the constant penalty assumption and explicitly enforces the inter-scanline consistency constraint. The resulting algorithm can handle sequences that contain large motions and can produce optical flows with 100% density over the entire image domain. The experimental results indicate that it can generate more accurate optical flows than existing approaches.     

视频微小运动放大的加速方法

通过放大视频中的微小运动,可以展现一些肉眼不可见的运动信息。原始欧拉视频放大技术（EVM方法）将输入的视频由RGB颜色空间转换为YIQ颜色空间,对视频的YIQ三个通道进行运动放大处理,这样的处理过程十分耗时。在原方法的基础上,分别提出了基于RGB颜色空间和基于Y通道的运动放大加速方法。将Y通道方法、RGB方法和EVM方法进行实验对比,结果表明：RGB方法相对EVM方法在处理速度方面有所提升,但是运动放大效果差别较大;Y通道方法相对于EVM方法在处理速度方面得到了明显的提升,而且在放大效果方面基本保持一致。     

Multiscale motion saliency for keyframe extraction from motion capture sequences

Motion capture is an increasingly popular animation technique; however data acquired by motion capture can become substantial. This makes it difficult to use motion capture data in a number of applications, such as motion editing, motion understanding, automatic motion summarization, motion thumbnail generation, or motion database search and retrieval. To overcome this limitation, we propose an automatic approach to extract keyframes from a motion capture sequence. We treat the input sequence as motion curves, and obtain the most salient parts of these curves using a new proposed metric, called 'motion saliency'. We select the curves to be analysed by a dimension reduction technique, Principal Component Analysis (PCA). We then apply frame reduction techniques to extract the most important frames as keyframes of the motion. With this approach, around 8% of the frames are selected to be keyframes for motion capture sequences. Copyright © 2010 John Wiley & Sons, Ltd.     

An interactive data-driven driving simulator using motion blending

Compared to the motion equations the data-driven method can simulate reality from sampling of real motions but real-time interaction between a user and the simulator is problematic. Existing data-driven motion generation methods simply record and replay the motion of the vehicle. Character animation technology enables a user to control motions that are generated by a motion capture database and an appropriate motion control algorithm. We propose a data-driven motion generation method and implement a driving simulator by adapting the method of motion capture. The motion data sampled from a real vehicle are transformed into appropriate data structures called motion blocks, and then a series of motion blocks are saved into the motion database. During simulation, the driving simulator searches for and synthesizes optimal motion blocks from the motion database and generates motion streams that reflect the current simulation conditions and parameterized user demands. We demonstrate the proposed method through experiments with the driving simulator.     

Multiframe Motion Segmentation with Missing Data Using PowerFactorization and GPCA

We consider the problem of segmenting multiple rigid-body motions from point correspondences in multiple affine views. We cast this problem as a subspace clustering problem in which point trajectories associated with each motion live in a linear subspace of dimension two, three or four. Our algorithm involves projecting all point trajectories onto a 5-dimensional subspace using the SVD, the PowerFactorization method, or RANSAC, and fitting multiple linear subspaces representing different rigid-body motions to the points in ℝ⁵ using GPCA. Unlike previous work, our approach does not restrict the motion subspaces to be four-dimensional and independent. Instead, it deals gracefully with all the spectrum of possible affine motions: from two-dimensional and partially dependent to four-dimensional and fully independent. Our algorithm can handle the case of missing data, meaning that point tracks do not have to be visible in all images, by using the PowerFactorization method to project the data. In addition, our method can handle outlying trajectories by using RANSAC to perform the projection. We compare our approach to other methods on a database of 167 motion sequences with full motions, independent motions, degenerate motions, partially dependent motions, missing data, outliers, etc. On motion sequences with complete data our method achieves a misclassification error of less that 5% for two motions and 29% for three motions.