Skeleton-based Methods Research Articles

Siamese denoising autoencoders for joints trajectories reconstruction and robust gait recognition

Dynamics of body skeletons convey significant information for human gait recognition. However, it is inevitable that missing points, overlapping error, or confusion of left and right error will frequently occur during the process of skeleton estimation. Existing skeleton-based methods have difficulty in achieving satisfactory performance in gait recognition since they treat the noisy data and the normal data equally to the recognition process. In this paper, we propose a novel skeleton-based model called Siamese Denoising Autoencoder networks (Siamese DAE), which can automatically learn to remove position noise, recover missing skeleton points and correct outliers in joint trajectories. More precisely, we construct an encoder that compresses the characteristics of input trajectories into a latent space and a decoder that attempts to reconstruct more accurate skeleton trajectories from the latent feature. The corrected joint trajectories not only lead to higher discriminative power but also stronger generalization capability. Moreover, we design a Siamese structure to reduce intra-class variations and increase inter-class variations of the encoded features. Experiments demonstrate that our method enhances the robustness against inaccurate skeleton estimation and achieves substantial improvements over mainstream skeleton-based methods for gait recognition.

Deforming generalized cylinders without self-intersection by means of a parametric center curve

Large-scale deformations of a tubular object, or generalized cylinder, are often defined by a target shape for its center curve, typically using a parametric target curve. This task is non-trivial for free-form deformations or direct manipulation methods because it is hard to manually control the centerline by adjusting control points. Most skeleton-based methods are no better, again due to the small number of manually adjusted control points. In this paper, we propose a method to deform a generalized cylinder based on its skeleton composed of a centerline and orthogonal cross sections. Although we are not the first to use such a skeleton, we propose a novel skeletonization method that tries to minimize the number of intersections between neighboring cross sections by means of a relative curvature condition to detect intersections. The mesh deformation is first defined geometrically by deforming the centerline and mapping the cross sections. Rotation minimizing frames are used during mapping to control twisting. Secondly, given displacements on the cross sections, the deformation is decomposed into finely subdivided regions. We limit distortion at these vertices by minimizing an elastic thin shell bending energy, in linear time. Our method can handle complicated generalized cylinders such as the human colon.

A High Invariance Motion Representation for Skeleton-Based Action Recognition

Human action recognition is very important and significant research work in numerous fields of science, for example, human–computer interaction, computer vision and crime analysis. In recent years, relative geometry features have been widely applied to the description of relative relation of body motion. It brings many benefits to action recognition such as clear description, abundant features etc. But the obvious disadvantage is that the extracted features severely rely on the local coordinate system. It is difficult to find a bijection between relative geometry and skeleton motion. To overcome this problem, many previous methods use relative rotation and translation between all skeleton pairs to increase robustness. In this paper we present a new motion representation method. It establishes a motion model based on the relative geometry with the aid of special orthogonal group SO(3). At the same time, we proved that this motion representation method can establish a bijection between relative geometry and motion of skeleton pairs. After the motion representation method in this paper is used, the computation cost of action recognition reduces from the two-way relative motion (motion from A to B and B to A) to one-way relative motion (motion from A to B or B to A) between any skeleton pair, namely, permutation problem [Formula: see text] is simplified into combinatorics problem [Formula: see text]. Finally, the experimental results of the three motion datasets are all superior to present skeleton-based action recognition methods.

Skeleton based Non rigid Shape Matching

This paper offers a new feature for matching deformable objects based on skeleton. Effectiveness of the skeleton-based methods depends to a large extent on the quality and perfection of the extracted skeleton. When the skeleton is extracted, additional branches of which, with the shape contour segmentation, are omitted by DCE method. The pruned skeleton is robust to the complex deformations and noise. After the pruned skeleton is obtained, minimum distance of contour points of the shape from the skeleton is calculated. This distance almost is constant against the articulated changes which this feature is used for finding the corresponding points. In order to carry out matching, the suggested method is applied on the MPEG7 dataset. Results show that the proposed algorithm is highly effective for matching objects.

Shape-aware skeletal deformation for 2D characters

This paper presents a skeleton-based method for deforming 2D characters. While previous skeleton-based methods drive the shape deformation by binding the skeleton to the shape, our method does so b...

Research and Perspective on Shape Matching

摘要: 形状匹配及分类是计算机视觉中的重要问题. 近年来,以形状上下文为代表的基于轮廓的形状匹配方法和以奇点图为代表的基于骨架的形状匹配方法获得了长足的发展. 本文介绍了形状匹配问题的基本概念, 分析了形状匹配问题的难点, 按照基于轮廓和基于骨架的分类方法对近年来最新出现的形状表示与形状匹配的方法进行了详尽的介绍, 并介绍了基于度量学习的形状检索方法, 本文还详细介绍了近年来形状匹配研究领域常用的一些测试数据库, 之后对局部形状匹配和形状分类等有潜力的研究方向进行了展望. 最后对形状匹配的整体框架及其应用前景进行了总结. 关键词: 形状表示 / 形状匹配 / 形状分类 / 轮廓 / 骨架

Pose Space Surface Manipulation

Example-based mesh deformation techniques produce natural and realistic shapes by learning the space of deformations from examples. However, skeleton-based methods cannot manipulate a global mesh structure naturally, whereas the mesh-based approaches based on a translational control do not allow the user to edit a local mesh structure intuitively. This paper presents an example-driven mesh editing framework that achieves both global and local pose manipulations. The proposed system is built with a surface deformation method based on a two-step linear optimization technique and achieves direct manipulations of a model surface using translational and rotational controls. With the translational control, the user can create a model in natural poses easily. The rotational control can adjust the local pose intuitively by bending and twisting. We encode example deformations with a rotation-invariant mesh representation which handles large rotations in examples. To incorporate example deformations, we infer a pose from the handle translations/rotations and perform pose space interpolation, thereby avoiding involved nonlinear optimization. With the two-step linear approach combined with the proposed multiresolution deformation method, we can edit models at interactive rates without losing important deformation effects such as muscle bulging.

Shape Deformation Using a Skeleton to Drive Simplex Transformations

This paper presents a novel skeleton-based method for deforming meshes (using an approximate skeleton, rather than a precise medial axis). The significant difference from previous skeleton-based methods is that the latter use the skeleton to control movement of vertices whereas we use it to control the simplices defining the model. By doing so, errors that occur near joints in other methods can be spread over the whole mesh, using an optimization process, resulting in smooth transitions near joints of the skeleton. By controlling simplices, our method has the advantage that no vertex weights need to be defined on the bones, which is a tedious requirement in previous skeleton-based methods. Our method can also easily be extended to control deformation by moving a few chosen line segments or vertices embedded in the object, rather than a skeleton.