Learning cooperative dynamic manipulation skills from human demonstration videos

This article proposes a method for learning and robotic replication of dynamic collaborative tasks from offline videos. The objective is to extend the concept of learning from demonstration (LfD) to dynamic scenarios, benefiting from widely available or easily producible offline videos. To achieve this goal, we decode important dynamic information, such as the Configuration Dependent Stiffness (CDS), which reveals the contribution of arm pose to the arm endpoint stiffness, from a three-dimensional human skeleton model. Next, through encoding of the CDS via Gaussian Mixture Model (GMM) and decoding via Gaussian Mixture Regression (GMR), the robot’s Cartesian impedance profile is estimated and replicated. We demonstrate the proposed method in a collaborative sawing task with leader–follower structure, considering environmental constraints and dynamic uncertainties. The experimental setup includes two Panda robots, which replicate the leader–follower roles and the impedance profiles extracted from a two-persons sawing video.     

XOR-based visual cryptography scheme with essential shadows

Visual cryptography scheme with essential shadows (EVCS) is of great significance since it provides different levels of the importance to shadows. In this paper, we propose a general construction method for (t, s, k, n)-VCS with essential shadows based on XOR operation ((t, s, k, n)-EXVCS), which originates from the partition of access structure. The secret image is encrypted into s essential shadows and n-s non-essential shadows. Any k shadows including at least t essentials can cooperate to decode the secret image and the decoding process is implemented by XOR operation on the involved shadows. Our scheme achieves perfectly reconstruction of secret image in the revealed image and the less size of shadows and revealed images. The experiments are conducted to testify the feasibility and practicability of the proposed scheme.     

An enhanced image quality assessment by synergizing superpixels and visual saliency

Superpixel and saliency-based evaluation methods play important roles in full reference image quality assessment (FR IQA). However, we find that these methods have one complementary principle and three limitations: (1) the weighted maps of superpixel-based methods conflict with the perception of the human visual system; (2) saliency-based methods are inefficient in terms of the block distortion; (3) the general two-direction gradient extraction factor must be extended to be multidirectional. To address these limitations, we propose an enhanced image quality assessment by synergizing superpixels and visual saliency. Specifically, the calculation of a newly proposed framework involves three similarities and two strategies: the saliency, superpixel and multidirectional gradient similarities of the neighborhoods, and the saliency pooling strategy, the fusion strategy of these similarities. Theoretical analysis and experimental results show that the proposed method can effectively address the limitations noted above and outperform the existing methods.     

Fantastic plastic? An image-based test method to detect aesthetic defects in batches based on reference samples

The production of high-end manufactured products requires Aesthetic Quality Control (AQC) in the form of human visual inspection. Manufactures can reduce AQC costs by incorporating semi-automated visual defect detection in units with the existing 3D metrology scans. This paper demonstrates how an image-based test method for defect detection can reduce the workload related to human visual inspection by proposing a median master comparison of batch image series. Our contribution consist of a) contrast enhancing and sorting batch image series for human visual inspection and b) providing a quality index (nQI) incorporated into statistical process control (SPC) for monitoring and controlling the AQC process. Our data shows that the median master differencing together with the nQI is great for classification of defects in batch images series. We introduce a SPC design proposal where individual batches as well as aggregated data can be inspected in synergy with the principles of Six Sigma. Based on Six Sigma control limits we have reduced the number of images in need of review by AQC assessors by a factor of 13.     

Visual analysis of speed bumps using floating car dataset

The analysis of traffic movements in urban populations is important for safety measures to control speed when the presence of pedestrians is high. Most of the studies in the traffic domain are directed towards statistical analyses of speed, also to measure the physical parameters of bumps on roads, and effect of speed-calming measures. However, less is known about the effectiveness of varying intervals between speed-calming measures and their influence on the driving behaviour of vehicles from a spatio-temporal perspective. To fill this gap and understand these aspects, in this research we propose visual analytics techniques. To explore the influence of the distance between bumps on speed change behaviour of vehicles in relation to the permitted speed limit, several use case studies were selected with the close cooperation of a traffic expert. The aim was first to study the influence of the speed bumps on speeding, and then establish any connection between speed bump intervals and vehicle speeding in the selected use cases. The results of our investigation did reveal minor differences and influence of a time aspect on traffic volume and speed development. Also, differences in actual speed of the vehicles in relation to the distance of installed bumps were detected.     

基于纳米金粒子可视化分析检测的研究进展

基于纳米金粒子表面等离子共振性质发展起来的可视化分析,由于具有灵敏度高、可设计性强、分析过程快速等优点已被广泛地应用于各类目标分析物的检测,成为一种极具潜力的分析手段。本文综述了基于纳米金粒子表面等离子共振可视化分析检测的研究进展,并对其发展趋势进行了展望。     

Towards Artificial Visual Sensory System: Organic Optoelectronic Synaptic Materials and Devices

Developing an artificial visual sensory system requires optoelectronic materials and devices that can mimic the behavior of biological synapses. Organic/polymeric semiconductors have emerged as promising candidates for optoelectronic synapses due to their tunable optoelectronic properties, mechanic flexibility, and biological compatibility. In this review, we discuss the recent progress in organic optoelectronic synaptic materials and devices, including their design principles, working mechanisms, and applications. We also highlight the challenges and opportunities in this field and provide insights into potential applications of these materials and devices in next-generation artificial visual systems. By leveraging the advances in organic optoelectronic materials and devices, we can envision its future development in artificial intelligence.     

低功耗电润湿显示器视频显示驱动系统设计

针对电润湿显示器在播放视频时功耗较大的问题,设计了一套基于人眼视觉特性的低功耗视频显示驱动系统。该系统采用直方图修正与直方图匹配的方法,结合人眼视觉特性设定阈值,减少显示图像中高亮像素的数量从而降低系统功耗;同时提出一种低功耗多灰度驱动波形,结合电润湿显示器的光电特性,在实现灰度精准调制的同时进一步降低系统的功耗。实验结果表明:该系统改善了油墨分裂、油墨回流等问题,成功驱动了1024×768分辨率的电润湿显示器且视频显示效果良好,像素的最高灰度等级达到16阶,在测量同一段视频时,平均功率降低至1.75 W,功耗减少约20%.     

Discriminative two-level feature selection for realistic human action recognition

Constructing the bag-of-features model from Space–time interest points (STIPs) has been successfully utilized for human action recognition. However, how to eliminate a large number of irrelevant STIPs for representing a specific action in realistic scenarios as well as how to select discriminative codewords for effective bag-of-features model still need to be further investigated. In this paper, we propose to select more representative codewords based on our pruned interest points algorithm so as to reduce computational cost as well as improve recognition performance. By taking human perception into account, attention based saliency map is employed to choose salient interest points which fall into salient regions, since visual saliency can provide strong evidence for the location of acting subjects. After salient interest points are identified, each human action is represented with the bag-of-features model. In order to obtain more discriminative codewords, an unsupervised codeword selection algorithm is utilized. Finally, the Support Vector Machine (SVM) method is employed to perform human action recognition. Comprehensive experimental results on the widely used and challenging Hollywood-2 Human Action (HOHA-2) dataset and YouTube dataset demonstrate that our proposed method is computationally efficient while achieving improved performance in recognizing realistic human actions.