Detection of sudden pedestrian crossings for driving assistance systems

In this paper, we study the problem of detecting sudden pedestrian crossings to assist drivers in avoiding accidents. This application has two major requirements: to detect crossing pedestrians as early as possible just as they enter the view of the car-mounted camera and to maintain a false alarm rate as low as possible for practical purposes. Although many current sliding-window-based approaches using various features and classification algorithms have been proposed for image-/video-based pedestrian detection, their performance in terms of accuracy and processing speed falls far short of practical application requirements. To address this problem, we propose a three-level coarse-to-fine video-based framework that detects partially visible pedestrians just as they enter the camera view, with low false alarm rate and high speed. The framework is tested on a new collection of high-resolution videos captured from a moving vehicle and yields a performance better than that of state-of-the-art pedestrian detection while running at a frame rate of 55 fps.     

A self-learning particle swarm optimizer for global optimization problems

Particle swarm optimization (PSO) has been shown as an effective tool for solving global optimization problems. So far, most PSO algorithms use a single learning pattern for all particles, which means that all particles in a swarm use the same strategy. This monotonic learning pattern may cause the lack of intelligence for a particular particle, which makes it unable to deal with different complex situations. This paper presents a novel algorithm, called self-learning particle swarm optimizer (SLPSO), for global optimization problems. In SLPSO, each particle has a set of four strategies to cope with different situations in the search space. The cooperation of the four strategies is implemented by an adaptive learning framework at the individual level, which can enable a particle to choose the optimal strategy according to its own local fitness landscape. The experimental study on a set of 45 test functions and two real-world problems show that SLPSO has a superior performance in comparison with several other peer algorithms.     

Example-based automatic music-driven conventional dance motion synthesis

We introduce a novel method for synthesizing dance motions that follow the emotions and contents of a piece of music. Our method employs a learning-based approach to model the music to motion mapping relationship embodied in example dance motions along with those motions' accompanying background music. A key step in our method is to train a music to motion matching quality rating function through learning the music to motion mapping relationship exhibited in synchronized music and dance motion data, which were captured from professional human dance performance. To generate an optimal sequence of dance motion segments to match with a piece of music, we introduce a constraint-based dynamic programming procedure. This procedure considers both music to motion matching quality and visual smoothness of a resultant dance motion sequence. We also introduce a two-way evaluation strategy, coupled with a GPU-based implementation, through which we can execute the dynamic programming process in parallel, resulting in significant speedup. To evaluate the effectiveness of our method, we quantitatively compare the dance motions synthesized by our method with motion synthesis results by several peer methods using the motions captured from professional human dancers' performance as the gold standard. We also conducted several medium-scale user studies to explore how perceptually our dance motion synthesis method can outperform existing methods in synthesizing dance motions to match with a piece of music. These user studies produced very positive results on our music-driven dance motion synthesis experiments for several Asian dance genres, confirming the advantages of our method.     

The right view from the wrong location: depth perception in stereoscopic multi-user virtual environments

Stereoscopic depth cues improve depth perception and increase immersion within virtual environments (VEs). However, improper display of these cues can distort perceived distances and directions. Consider a multi-user VE, where all users view identical stereoscopic images regardless of physical location. In this scenario, cues are typically customized for one "leader" equipped with a head-tracking device. This user stands at the center of projection (CoP) and all other users ("followers") view the scene from other locations and receive improper depth cues. This paper examines perceived depth distortion when viewing stereoscopic VEs from follower perspectives and the impact of these distortions on collaborative spatial judgments. Pairs of participants made collaborative depth judgments of virtual shapes viewed from the CoP or after displacement forward or backward. Forward and backward displacement caused perceived depth compression and expansion, respectively, with greater compression than expansion. Furthermore, distortion was less than predicted by a ray-intersection model of stereo geometry. Collaboration times were significantly longer when participants stood at different locations compared to the same location, and increased with greater perceived depth discrepancy between the two viewing locations. These findings advance our understanding of spatial distortions in multi-user VEs, and suggest a strategy for reducing distortion.     

Ultrasonic phased-array scanner with digital echo synthesis for Doppler echocardiography

For the purpose of the quantitative assessment of subtle disease processes in the cardiovascular system an electronically steered sector scanner that combines echographic imaging and Doppler blood velocity measurements has been developed. The integrated operation of a fast Fourier transform (FFT) Doppler signal processor for the simultaneous blood velocity evaluation of 64 individual gates is among the specific design goals. The instrument incorporates an unusually high degree of digital signal processing, which allows for high integration density, easy manufacturing and high reliability in future designs. The complex Doppler spectra are determined for each of the 64 Doppler gates in real time, and the subsequent computation of the first moment provides a reliable estimate of the mean blood flow velocities at the respective locations. The instantaneous velocity profile along the Doppler beam is displayed together with the calculated volume flow rate and a range-selected complete frequency spectrum. Results of both in vitro and in vivo tests indicate that in the future, a higher degree of digital signal processing could be implemented in complex ultrasonic systems.     

Small blaze angle gratings with various surface treatments for use in the extreme ultraviolet

Three replica gratings were made from a single master; two were epoxy overcoated with different thicknesses of gold, and one was replicated directly in gold. Their performance in EUV light at grazing incidence angles was measured and compared.     

非饱和水力传导系数的简易实验室测量方法

本文描述一种采用常规的实验装置在短时间内测量土层非饱和水力传导率的简易方法,称为"向上入渗法".这一方法基于在控制压力的条件下测量土的吸水量,从而来估算土层在湿润状态下的水力传导系数.将其结果与恒定加水量法、小增量压力变化法、一步法以及杰克逊理论模型的计算结果进行了比较.采用向上入渗法与采用恒定加水量法的结果很接近,但稍大于一步法而小于杰克逊理论模型的结果.这种测定方法限定负压值约为一米.