The dangerous driving action seriously affects traffic safety and may cause severe road disasters. Dangerous driving action recognition in Internet of Vehicles (IOV) has been widely exploited to reduce traffic accident risks by discovering and then transmitting the recognition results to autonomous machines or other vehicles. General action recognition models using deep learning networks usually have a large number of parameters and require a significant amount of memory and computational power, which cannot meet the lightweight and real-time requirements of action recognition models deployed on resource-limited onboard devices, e.g., vehicles. As a result, we introduce the broad learning system (BLS) into onboard dangerous driving action recognition tasks, classify actions based on skeleton data, and design the graph convolutional representation (GCR) algorithm and graph convolutional broad learning system (GCBLS) classification model to speed up the recognition process. We verified through ablation experiments that the GCR algorithm can effectively represent skeleton data and improve classification accuracy datasets. In addition, comparative experiments in State Farm and Driver Skeleton datasets show that the GCBLS model has the characteristics of lightweight, real-time, and high accuracy. We also designed a workflow for "noise" caused by poor pose estimations in practical applications, then deployed the proposed workflow on onboard devices, which can run at speeds above 27 FPS with high accuracy, which proves the effectiveness and practicability of our proposed algorithms and models for IOV.

Jianqiao Li, Chunyuan Li, Guoyin Wang, Hao Fu, Yuhchen Lin, Liqun Chen, Yizhe Zhang, Chenyang Tao, Ruiyi Zhang, Wenlin Wang, Dinghan Shen, Qian Yang, Lawrence Carin. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

As a fundamental issue in lifelong learning, catastrophic forgetting is directly caused by inaccessible historical data; accordingly, if the data (information) were memorized perfectly, no forgetting should be expected. Motivated by that, we propose a GAN memory for lifelong learning, which is capable of remembering a stream of datasets via generative processes, with \emph{no} forgetting. Our GAN memory is based on recognizing that one can modulate the ``style'' of a GAN model to form perceptually-distant targeted generation. Accordingly, we propose to do sequential style modulations atop a well-behaved base GAN model, to form sequential targeted generative models, while simultaneously benefiting from the transferred base knowledge. The GAN memory -- that is motivated by lifelong learning -- is therefore itself manifested by a form of lifelong learning, via forward transfer and modulation of information from prior tasks. Experiments demonstrate the superiority of our method over existing approaches and its effectiveness in alleviating catastrophic forgetting for lifelong classification problems. Code is available at \url{https://github.com/MiaoyunZhao/GANmemory_LifelongLearning}.

Fundamental aspects of adversarial learning are investigated, with learning based on samples from the target distribution (conventional GAN setup). With insights so garnered, adversarial learning is extended to the case for which one has access to an unnormalized form $u(x)$ of the target density function, but no samples. Further, new concepts in GAN regularization are developed, based on learning from samples or from $u(x)$. The proposed method is compared to alternative approaches, with encouraging results demonstrated across a range of applications, including deep soft Q-learning.

The addition of manipulation capabilities to telepresence robots holds the promise of enabling remote humans to perform tele-labor, hands-on training, and collaborative manipulation, but the use of a robot as a mediator to humanhuman physical interaction is not yet well understood. This paper studies the impact of telepresence modalities in the context of robot-mediated object handover. A teleoperation system was developed involving a bimanual mobile manipulator with telepresence head and sensing capabilities, and a user study was conducted with n=10 pairs of subjects under a variety of audio and visual telepresence conditions. Results show that telepresence does not significantly affect objective handover fluency, but both audio and video telepresence do significantly improve user experience on subjective measures including intimacy and perceived fluency.

In this paper, we investigate how the recently emerged photography technology--the light field--can benefit depth map estimation, a challenging computer vision problem. A novel framework is proposed to reconstruct continuous depth maps from light field data. Unlike many traditional methods for the stereo matching problem, the proposed method does not need to quantize the depth range. By making use of the structure information amongst the densely sampled views in light field data, we can obtain dense and relatively reliable local estimations. Starting from initial estimations, we go on to propose an optimization method based on solving a sparse linear system iteratively with a conjugate gradient method. Two different affinity matrices for the linear system are employed to balance the efficiency and quality of the optimization. Then, a depth-assisted segmentation method is introduced so that different segments can employ different affinity matrices. Experiment results on both synthetic and real light fields demonstrate that our continuous results are more accurate, efficient, and able to preserve more details compared with discrete approaches.

Light field analysis recently received growing interest, since its rich structure information benefits many computer vision tasks. This paper presents a novel method to reconstruct continuous depth maps from light field data. Conventional approaches usually treat depth map reconstruction as an optimization problem with discrete labels. On the contrary, our proposed method can obtain continuous depth maps by solving a linear system, which preserves richer details compared with conventional discrete approaches. Structure tensor is employed to extract raw depth information and corresponding confidence levels from the light field data. We introduce a method to reduce the adverse effect of unreliable local estimations, which helps to get rid of errors in specular areas and edges where depth values are discontinuous. Experiments on both synthetic and real light field data demonstrate the effectiveness of the proposed method.