In this paper, we propose a Point Fractal Network (PF-Net), a novel learning-based approach for precise and high-fidelity point cloud completion. Unlike existing point cloud completion networks, which generate the overall shape of the point cloud from the incomplete point cloud and always change existing points and encounter noise and geometrical loss, PF-Net preserves the spatial arrangements of the incomplete point cloud and can figure out the detailed geometrical structure of the missing region(s) in the prediction. To succeed at this task, PF-Net estimates the missing point cloud hierarchically by utilizing a feature-points-based multi-scale generating network. Further, we add up multi-stage completion loss and adversarial loss to generate more realistic missing region(s). The adversarial loss can better tackle multiple modes in the prediction. Our experiments demonstrate the effectiveness of our method for several challenging point cloud completion tasks.

As a concise framework, Mask R-CNN achieves promising performance in object detection and instance segmentation. However, there is room for improvement in two aspects. One is that performing multi-task prediction needs more credible feature extraction and multi-scale features integration to handle objects with varied scales. We address this problem by using a novel neck module called SA-FPN (Scale Aware Feature Pyramid Networks), which can accurately help detect and segment the objects of multiple scales. The other is that the isolation between detection and instance segmentation branch exists, causing the gap between training and testing processes. So we propose a unified head module named EJ-Head (Effective Joint Head) to combine two branches into one head, not only realizing the interaction between two tasks, but also enhancing the effectiveness of multi-task learning. Comprehensive experiments on MS-COCO benchmark show that our proposed methods bring noticeable gains for both object detection and instance segmentation.1

This article is to develop virtual reality-based framework for enhancing walking abilities using exoskeleton robotics. On one hand, China will face an ever-increasing and unprecedented demand for rehabilitation treatment and training for a long period of time in the future; on the other hand, backward and shortage of rehabilitation training resources urgently need to be rapidly upgraded through scientific and technological progress. A lightweight and intelligent system for indoor walking abilities rehabilitation training that combines virtual reality and exoskeleton robotics will effectively improve the rehabilitation training conditions, efficiency and effects through scientific and technological means. Through in-depth analysis and tracking of the training plan, combined with the research results of gait theory, the article describes a customized exoskeleton with VR glass and local cloud, and the sensing process, virtual scene establishment and presentation, human-robot interaction mechanism, virtual training and performance evaluation, and constraints are elaborated. Large-scale system testing and data collection and analysis, and the heterogeneity of rehabilitation training with this method will be the focus of future research.

State-of-the-art hand gesture recognition methods have investigated the spatiotemporal features based on 3D convolutional neural networks (3DCNNs) or convolutional long short-term memory (ConvLSTM). However, they often suffer from the inefficiency due to the high computational complexity of their network structures. In this paper, we focus instead on the 1D convolutional neural networks and propose a simple and efficient architectural unit, Multi-Kernel Temporal Block (MKTB), that models the multi-scale temporal responses by explicitly applying different temporal kernels. Then, we present a Global Refinement Block (GRB), which is an attention module for shaping the global temporal features based on the cross-channel similarity. By incorporating the MKTB and GRB, our architecture can effectively explore the spatiotemporal features within tolerable computational cost. Extensive experiments conducted on public datasets demonstrate that our proposed model achieves the state-of-the-art with higher efficiency. Moreover, the proposed MKTB and GRB are plug-and-play modules and the experiments on other tasks, like video understanding and video-based person re-identification, also display their good performance in efficiency and capability of generalization.

The speech enhancement problem is studied using an improved robust generalized sidelobe canceler (GSC) beamforming algorithm based on microphone array, in the cases of speaker noise and the music interferences. The conventional GSC algorithm based on variable step size and a priori signal-to-noise ratio (SNR) algorithm is not robust under the nonstationary noise because the solution of the signal-to-noise ratio (SNR) is not given. To enhance the robustness, in this paper, a improved GSC algorithm is developed, where adaptive filter coefficients are updated based on signal output power ratio (SPR). The numerical studies including speaker noise and music noise demonstrate that the improved algorithm outperforms the traditional GSC and the GSC based on variable step size technique.

Leaf identification is a significant and challenging task for computer vision community due to the large intra-class variations and small inter-class differences of leaf image patterns. In this paper, a novel shape descriptor, called Integral Contour Angle (ICA), is proposed for accurate classification and retrieval of leaf images. For a contour point, two bunches of vectors emanate from it and ends at its left neighbor contour points and its right neighbor ones respectively. Their average vectors form an angle (termed ICA) having inherent invariance to translation, rotation and scaling of leaf shape. Changing the size of the neighborhood of the contour point naturally yields a set of ICAs at different scales. Group them to form a multiscale descriptor (called mICA) for the contour point. Collect the mICAs of all the contour points to construct a mICA set. The dissimilarity between two leaf shapes is measured by calculating the enhanced Hausdorff distance between their mICA sets. The experiment results on two popular leaf image datasets indicate that the proposed method outperforms the state-of-the-art methods.

This paper presents the development of a new type of robot capable of vertical and directional jumping. The robot uses soft silicone elastomer based pneumatic actuators as legs that accelerate the platform upwards by rapid pressurization. The robot is able to control and adjust the direction of the jumping by altering the timing patterns in which the individual legs are activated.