3D environment artists typically use 2D screens and 3D modeling software to achieve their creation. However, creating 3D content using 2D tools is counterintuitive. Moreover, the process would be inefficient for junior artists in the absence of a reference. We develop a system called GetWild, which employs artificial intelligence (AI) models to generate the prototype of 3D objects/terrain and allows users to further edit the generated content in the virtual space. With the aid of AI, the user can capture an image to obtain a rough 3D object model, or start with drawing simple sketches representing the river, the mountain peak and the mountain ridge to create a 3D terrain prototype. Further, the virtual reality (VR) technique is used to provide an immersive design environment and intuitive interaction (such as painting, sculpturing, coloring, and transformation) for users to edit the generated prototypes. Compared with the existing 3D modeling software and systems, the proposed VR editing system with AI-generated 3D objects/terrain provides a more efficient way for the user to create virtual artwork.

This paper presents a monolithic optical encoder using CMOS image sensor (CIS) with background light cancellation. Both the absolute and incremental encoders are designed and fabricated on the same chip with dual photodiode arrays and the corresponding readout circuits. The absolute encoder is implemented using 42 columns of pixel with adjustable dual-threshold quantizer. The incremental encoder is implemented using the developed common-centroid interlacing (CCI) photodiode (PD) arrangement for the sensing of four quadrature sinusoidal signal with a 90° phase shift to each other. A fully-differential transimpedance amplifier (FDTIA) with current-mode subtraction (CMS) is proposed to achieve the common-mode signal rejection and background light cancellation. With the implemented CCI PD array, CMS FDTIA, programmable gain amplifiers (PGA), and 12-b SAR ADCs, the prototype achieves a SNR of 60dB, a total power of 10mW, and the maximum displacement error of 0.22μm. Compared to the reported work [5], it achieves a SNR improvement of 15dB, a 2× power reduction, and a 2× accuracy.

Emotion reacts human beings' physiological and psychological status. Galvanic Skin Response (GSR) can reveal the electrical characteristics of human skin and is widely used to recognize the presence of emotion. In this work, we propose an emotion recognition frame-work based on deep hybrid neural networks, in which 1D CNN and Residual Bidirectional GRU are employed for time series data analysis. The experimental results show that the proposed method can outperform other state-of-the-art methods. In addition, we port the proposed emotion recognition model on Raspberry Pi and design a real-time emotion interaction robot to verify the efficiency of this work.

Generative adversarial network (GAN) has achieved impressive success on cross-domain generation, but it faces difficulty in cross-modal generation due to the lack of a common distribution between heterogeneous data. Most existing methods of conditional based cross-modal GANs adopt the strategy of one-directional transfer and have achieved preliminary success on text-to-image transfer. Instead of learning the transfer between different modalities, we aim to learn a synchronous latent space representing the cross-modal common concept. A novel network component named synchronizer is proposed in this work to judge whether the paired data is synchronous/corresponding or not, which can constrain the latent space of generators in the GANs. Our GAN model, named as SyncGAN, can successfully generate synchronous data (e.g., a pair of image and sound) from identical random noise. For transforming data from one modality to another, we recover the latent code by inverting the mappings of a generator and use it to generate data of different modality. In addition, the proposed model can achieve semi-supervised learning, which makes our model more flexible for practical applications.

While the content of virtual reality (VR) has grown explosively in recent years, the advance of designing user-friendly control interfaces in VR still remains a slow pace. The most commonly used device, such as gamepad or controller, has fixed shape and weight, and thus can not provide realistic haptic feedback when interacting with virtual objects in VR. In this work, we present a novel and lightweight tracking system in the context of manipulating handheld objects in VR. Specifically, our system can effortlessly synchronize the 3D pose of arbitrary handheld objects between the real world and VR in realtime performance. The tracking algorithm is simple, which delicately leverages the power of Leap Motion and IMU sensor to respectively track object's location and orientation. We demonstrate the effectiveness of our system with three VR applications use pencil, ping-pong paddle, and smartphone as control interfaces to provide users more immersive VR experience.

As databases can overlap each other, data matching that aims to identify data records or elements describing the same object is one of the fundamental problems in physical integration of databases. Matching results can be applied to induce more accurate and complete object descriptions, remove data redundancy, check data consistency and generate cross-links. In this paper, we present a multilevel approach to match data between pathway databases that consist of 3 levels of information, compounds, reactions and pathways. Compounds comprise reactions, and reactions comprise pathways. Our main idea is to use relationships discovered in lower levels to infer relationships in upper levels, and use relationships in upper levels to enhance relationships in lower levels. In particular, we first use attributes of compounds to identify compound matching, i.e. to map compounds in one database to those in another database. Compound matching is then used to induce reaction matching. Reaction matching is used to induce pathway matching as well as to enhance compound matching. We experiment our approach for integration of two well known pathway databases KEGG and MetaCyc. The experiment shows that our approach identifies 1025 matchings in compound level, 968 matchings in reaction level, and 387 relations in pathway level. According to matching result, we assign EC number to 84 reactions in MetaCyc that do not have EC number, and discover some duplicate errors in both databases. Furthermore, our approach identifies 532 of the 544 unification links provided by MetaCyc. The recall rate is 0.977.

In this paper, we exploit the concept of sharing GPS data for estimate traffic information. Explicitly, we implemented CarWeb, a platform to collect real-time GPS data from cars. Once receiving a sufficient amount of GPS data from cars, we are able to estimate traffic information (i.e., speed information of roads). As such, users are able to obtain the update-to-date traffic information in CarWeb platform. In addition, we propose two algorithms to estimate traffic information with GPS data given. A prototype of CarWeb is implemented and functionalities of each component in CarWeb are demonstrated in this paper.