Mitsubishi Electric

About: Mitsubishi Electric is a company organization based out in Ratingen, Germany. It is known for research contribution in the topics: Signal & Voltage. The organization has 23024 authors who have published 27591 publications receiving 255671 citations. The organization is also known as: Mitsubishi Electric Corporation & Mitsubishi Denki K.K..

ESPNet: End-to-end speech processing toolkit

Abstract: This paper introduces a new open source platform for end-to-end speech processing named ESPnet. ESPnet mainly focuses on end-to-end automatic speech recognition (ASR), and adopts widely-used dynamic neural network toolkits, Chainer and PyTorch, as a main deep learning engine. ESPnet also follows the Kaldi ASR toolkit style for data processing, feature extraction/format, and recipes to provide a complete setup for speech recognition and other speech processing experiments. This paper explains a major architecture of this software platform, several important functionalities, which differentiate ESPnet from other open source ASR toolkits, and experimental results with major ASR benchmarks.

FoldingNet: Point Cloud Auto-Encoder via Deep Grid Deformation

Abstract: Recent deep networks that directly handle points in a point set, e.g., PointNet, have been state-of-the-art for supervised learning tasks on point clouds such as classification and segmentation. In this work, a novel end-to-end deep auto-encoder is proposed to address unsupervised learning challenges on point clouds. On the encoder side, a graph-based enhancement is enforced to promote local structures on top of PointNet. Then, a novel folding-based decoder deforms a canonical 2D grid onto the underlying 3D object surface of a point cloud, achieving low reconstruction errors even for objects with delicate structures. The proposed decoder only uses about 7% parameters of a decoder with fully-connected neural networks, yet leads to a more discriminative representation that achieves higher linear SVM classification accuracy than the benchmark. In addition, the proposed decoder structure is shown, in theory, to be a generic architecture that is able to reconstruct an arbitrary point cloud from a 2D grid. Our code is available at http://www.merl.com/research/license#FoldingNet

Adaptively sampled distance fields: a general representation of shape for computer graphics

Abstract: Adaptively Sampled Distance Fields (ADFs) are a unifying representation of shape that integrate numerous concepts in computer graphics including the representation of geometry and volume data and a broad range of processing operations such as rendering, sculpting, level-of-detail management, surface offsetting, collision detection, and color gamut correction. Its structure is uncomplicated and direct, but is especially effective for quality reconstruction of complex shapes, e.g., artistic and organic forms, precision parts, volumes, high order functions, and fractals. We characterize one implementation of ADFs, illustrating its utility on two diverse applications: 1) artistic carving of fine detail, and 2) representing and rendering volume data and volumetric effects. Other applications are briefly presented.

Style machines

Abstract: We approach the problem of stylistic motion synthesis by learning motion patterns from a highly varied set of motion capture sequences. Each sequence may have a distinct choreography, performed in a distinct sytle. Learning identifies common choreographic elements across sequences, the different styles in which each element is performed, and a small number of stylistic degrees of freedom which span the many variations in the dataset. The learned model can synthesize novel motion data in any interpolation or extrapolation of styles. For example, it can convert novice ballet motions into the more graceful modern dance of an expert. The model can also be driven by video, by scripts or even by noise to generate new choreography and synthesize virtual motion-capture in many styles.