Parallel processing of motion and colour information

When the two eyes are confronted with sufficiently different versions of the visual environment, one or the other eye dominates perception in alternation. A similar situation may be created in the laboratory by presenting images to the left and right eyes which differ in orientation or colour. Although perception is dominated by one eye during rivalry, there are a number of instances in which visual processes nevertheless continue to integrate information from the suppressed eye. For example the interocular transfer of the motion after-effect is undiminished when induced during binocular rivalry. Thus motion information processing may occur in parallel with the rivalry process. Here we describe a novel example in which the visual system simultaneously exhibits binocular rivalry and vision that integrates signals from both eyes. This apparent contradiction is resolved by postulating parallel visual processes devoted to the analyses of colour and motion information. Counterphased gratings are viewed dichoptically such that for one eye the grating is composed of alternating yellow and black stripes (luminance) while for the other it is composed of alternating red and green stripes (chrominance). When the gratings are fused, a moving grating is perceived. A consistent direction of motion can only be achieved if left and right monocular signals are integrated by the nervous system. Yet the apparent colour of the binocular percept alternates between red-green and yellow-black. These observations demonstrate the segregation of processing by the early motion system from that affording the perception of colour. Although, in this stimulus, colour information in itself can play no part in the cyclopean perception of motion direction, colour is carried along perceptually (filled in) by the moving pattern which is integrated from both eyes.     

一种基于视觉信息的自主搬运机器人

提出了一种利用视觉信息进行定位和抓取的自主机器人搬运系统,此系统由带有两自由度操作手的轮式移动机器人和远程控制站组成。机器人利用视觉系统识别路标信息,采用MonteCarlo方法进行自定位。当靠近目标物体时,机器人利用视觉系统识别目标,并在利用该信息引导下自动抓取。由于机器人的运行是靠视觉系统获取的路标信息和目标物体信息引导的,同时具有遥操作控制功能,因此该系统能轻松完成复杂环境下搬运物体的任务。     

Neural correlates of perceptual motion coherence.

The motions of overlapping contours in a visual scene may arise from the physical motion(s) of either a single or multiple surface(s). A central problem facing the visual motion system is that of assigning the most likely interpretation. The rules underlying this perceptual decision can be explored using a visual stimulus formed by superimposing two moving gratings. The resultant percept is either that of a single coherently moving 'plaid pattern' (coherent motion) or of the two component gratings sliding noncoherently across one another (noncoherent motion). When plaid patterns are configured to mimic one transparent grating overlying another, the percept of noncoherent motion dominates. We now report that neurons in the visual cortex of rhesus monkeys exhibit changes in direction tuning that parallel this perceptual phenomenon: sensitivity to the motions of the component gratings is enhanced under conditions that favour the perception of noncoherent motion. These results challenge models of cortical visual processing that fail to take into account the contribution of figural image segmentation cues to the analysis of visual motion.     

Interocular rivalry revealed in the human cortical blind-spot representation

To understand conscious vision, scientists must elucidate how the brain selects specific visual signals for awareness. When different monocular patterns are presented to the two eyes, they rival for conscious expression such that only one monocular image is perceived at a time. Controversy surrounds whether this binocular rivalry reflects neural competition among pattern representations or monocular channels. Here we show that rivalry arises from interocular competition, using functional magnetic resonance imaging of activity in a monocular region of primary visual cortex corresponding to the blind spot. This cortical region greatly prefers stimulation of the ipsilateral eye to that of the blind-spot eye. Subjects reported their dominant percept while viewing rivalrous orthogonal gratings in the visual location corresponding to the blind spot and its surround. As predicted by interocular rivalry, the monocular blind-spot representation was activated when the ipsilateral grating became perceptually dominant and suppressed when the blind-spot grating became dominant. These responses were as large as those observed during actual alternations between the gratings, indicating that rivalry may be fully resolved in monocular visual cortex. Our findings provide the first physiological evidence, to our knowledge, that interocular competition mediates binocular rivalry, and indicate that V1 may be important in the selection and expression of conscious visual information.     

Dissociation between mental imagery and object recognition in a brain-damaged patient.

Visual imagery is the creation of mental representations that share many features with veridical visual percepts. Studies of normal and brain-damaged people reinforce the view that visual imagery and visual perception are mediated by a common neural substrate and activate the same representations. Thus, brain-damaged patients with intact vision who have an impairment in perception should have impaired visual imagery. Here we present evidence to the contrary from a patient with severely impaired object recognition (visual object agnosia) but with normal mental imagery. He draws objects in considerable detail from memory and uses information derived from mental images in a variety of tasks. In contrast, he cannot identify visually presented objects, even those he has drawn himself. He has normal visual acuity and intact perception of equally complex material in other domains. We conclude that rich internal representations can be activated to support visual imagery even when they cannot support visually mediated perception of objects.     

Visual control of action but not perception requires analytical processing of object shape

The visual perception of object shape depends on 'holistic' processing in which a given dimension cannot be perceptually isolated from the other dimensions of the object. The visual control of action (such as grasping an object), however, which is mediated by cortical areas that are largely independent of those mediating conscious perception, must take into account only the most action-relevant dimension of an object without being misled by other non-relevant object features. Here we report the results of two experiments showing that vision for perception and vision for action deal with objects in a fundamentally different manner. We tested participants' ability to make perceptual judgements of the width of different rectangular objects or to grasp them across their width, while in both cases ignoring length. Participants could not ignore length when making perceptual judgements of width but they could completely ignore length when grasping the same objects. These results suggest that in situations in which the elementary dimensions of an object's shape are perceived in a holistic manner, the same dimensions are treated analytically when a visually guided action is directed at that same object.     

Knowledge Enhanced Pre-Training Model for Vision-Language-Navigation Task

Vision-Language-Navigation(VLN) task is a cross-modality task that combines natural language processing and computer vision. This task requires the agent to automatically move to the destination according to the natural language instruction and the observed surrounding visual information. To make the best decision, in every step during the navigation, the agent should pay more attention to understanding the objects, the object attributes, and the object relationships. But most current methods process all received textual and visual information equally. Therefore, this paper integrates more detailed semantic connections between visual and textual information through three pre-training tasks(object prediction, object attributes prediction, and object relationship prediction). The model will learn better fusion representation and alignment between these two types of information to improve the success rate(SR) and generalization. The experiments show that compared with the former baseline models, the SR on the unseen validation set(Val Unseen) increased by 7%, and the SR weighted by path length(SPL) increased by 7%; the SR on the test set(Test) increased 4%, SPL increased by 3%.     

用单个摄像机获取三维视觉信息的一种方法

讨论了在平面坐标系上摄像机的标定和利用单个摄像机获取三维视觉信息的一种方法。实验结果表明，采用单个摄像机获取三维视觉信息的精度与摄像机相对目标物体的高度及对象物本身的厚度有关，在视场范围内，对象物的厚度越小，摄像机坐标中心相对于目标物体表面中心的距离越远，定位精度越好。     

感知模式对织物柔软性评价结果的影响

目前在织物柔软性的感官评价中,没有讨论感知模式(感觉模态和触摸方式)对评价结果的影响,导致评价结果间没有可比性,且评价方法难以标准化.通过让实验者在不同感觉模态(触觉单模态、视觉单模态和触视觉双模态)和触摸方式(捏持弯曲、弯曲按压和平置按压)下感官评价织物样品,对结果进行多元统计分析,讨论感知模式对织物柔软性感官估值的影响及不同感知模式之间的差异.研究表明,总体上感知模式对织物柔软性的影响不显著,但不同的触摸方式对较厚且压缩性好的织物影响显著;而对薄且弯曲刚度高的织物,柔软感估值在视觉单模态和触觉环境之间有显著差异.另外,就任一织物样品而言,全体实验者对其柔软感强度估值服从正态分布规律.研究结果为织物柔软感评价体系的完善提供依据.     

Efficient and Robust Feature Model for Visual Tracking

Long duration visual tracking of targets is quite challenging for computer vision,because the environments may be cluttered and distracting.Illumination variations and partial occlusions are two main difficulties in real world visual tracking.Existing methods based on hostile appearance information cannot solve these problems effectively.This paper proposes a feature-based dynamic tracking approach that can track objects with partial occlusions and varying illumination.The method represents the tracked o...     

一种用于力觉再现的柔性体变形仿真弹簧-质点模型

提出一种新型的在虚拟现实中可实现的弹簧-质点力/变形模型.它按质点连接顺序划分层次,以受力网格为第1层,构建类似同心环状的弹簧-质点系统.施力刚体以点接触的方式与柔性体进行力交互作用.运用虚功原理,结合施力刚体与受力柔性体组成的系统中虚拟力和质点位移的关系,区别于以往对局部变形的研究,对柔性体进行全局变形分析.柔性体变形是整个柔性体表面质点在力平衡条件下位移变化累加的结果.利用DELTA 手控器对柔性体的接触变形以及实时虚拟力反馈进行仿真,实验表明所提出的方法适用于柔性体的力反馈计算,满足精细作业对虚拟现实系统的要求.     

Large adjustments in visually guided reaching do not depend on vision of the hand or perception of target displacement

When we reach towards an object that suddenly appears in our peripheral visual field, not only does our arm extend towards the object, but our eyes, head and body also move in such a way that the image of the object falls on the fovea. Popular models of how reaching movements are programmed have argued that while the first part of the limb movement is ballistic, subsequent corrections to the trajectory are made on the basis of dynamic feedback about the relative positions of the hand and the target provided by central vision. These models have assumed that the adjustments are dependent on seeing the hand moving with respect to the target. Here we present evidence that a change in the position of a visual target during a reaching movement can modify the trajectory even when vision of the hand is prevented. Moreover, these dynamic corrections to the trajectory of the moving limb occur without the subject perceiving the change in target location. These findings demonstrate that visual feedback about the relative position of the hand and target is not necessary for visually driven corrections in reaching to occur, and the mechanisms that maintain the apparent stability of a target in space are dissociable from those that mediate the visuomotor output directed at that target.     

细胞神经网络生物视觉信息处理模型

分析了连续细胞神经网络,离散细胞神经网络和时滞细胞神经网络的数学模型和在图像处理方面的特点.根据人类视觉的产生机制,提出了一种由多个细胞神经网络组成的生物视觉信息处理模型.该模型充分利用了细胞神经网络的并行处理能力,具有快速性.本工作可作为机器人仿人视觉与视觉信息处理的研究基础.     

Neuronal correlate of pictorial short-term memory in the primate temporal cortex

It has been proposed that visual-memory traces are located in the temporal lobes of the cerebral cortex, as electric stimulation of this area in humans results in recall of imagery. Lesions in this area also affect recognition of an object after a delay in both humans and monkeys, indicating a role in short-term memory of images. Single-unit recordings from the temporal cortex have shown that some neurons continue to fire when one of two or four colours are to be remembered temporarily. But neuronal responses selective to specific complex objects, including hands and faces, cease soon after the offset of stimulus presentation. These results led to the question of whether any of these neurons could serve the memory of complex objects. We report here a group of shape-selective neurons in an anterior ventral part of the temporal cortex of monkeys that exhibited sustained activity during the delay period of a visual short-term memory task. The activity was highly selective for the pictorial information to be memorized and was independent of the physical attributes such as size, orientation, colour or position of the object. These observations show that the delay activity represents the short-term memory of the categorized percept of a picture.     

用Velocity Verlet积分器改进HMC抽样方法

Hamilton Monte Carlo (HMC) 方法是一种常用的快速抽样方法. 在对哈密尔顿方程进行抽样时,HMC方法使用Leapfrog积分器,这可能造成方程的位置和动量的迭代值在时间上不同步,产生的误差会降低抽样效率及抽样结果的稳定性. 为此,本文提出了IHMC（Improved HMC）方法,该方法用Velocity Verlet积分器代替Leapfrog积分器,在每次迭代时都计算两变量在同一时刻的值. 为验证方法的效果,本文进行了两个数值实验,一个是将方法应用于非对称随机波动率模型（RASV模型）的参数估计,另一个是将方法应用于方差伽马分布的抽样. 结果显示,IHMC方法比HMC方法的效率更高,结果更稳定.     

融合边缘语义信息的单目深度估计

单目深度估计研究是许多视觉任务的基础,从图像中得到边缘清晰,细节丰富的深度图对于后续任务具有重要的作用.针对当前单目深度估计模型中不能深度融合图像语义信息以及不能较好地利用图像对象的边缘信息问题,首先构建了超像素拓扑关系图,使用图神经网络提取局部边缘信息之间的相互关系,得到以超像素为节点的拓扑关系图,其次构建了基于编解...     

Cooperative Localization and Tracking in Distributed Robot-Sensor Networks

This paper addresses the problem of real-time position and orientation estimation of networked mobile robots in two-dimensional Euclidean space with simultaneous tracking of a rigid unknown object based on exteroceptive sensory information extracted from distributed vision systems. The sufficient and necessary conditions for team localization are proposed. A localization and object tracking approach based on statistical operators and graph searching algorithms is presented for a team of robots localized with heterogeneous sensors. The approach was implemented in an experimental platform consisting of car-like mobile robots equipped with omnidirectional video cameras and IEEE 802.1 lb wireless networking. The experimental results validate the approach.     

基于核心密度估计的动态目标分割改进模型

数字视频中的动态目标分割是基于计算机视觉技术的分析、识别系统中关键的处理步骤,分割结果的正确率决定了后期分析或识别过程的质量.在数字视频中存在各种有害动态像素,它们会降低分割过程的正确率.动态目标的阴影是这些有害像素的一种,因为它们本身不属于动态目标形状信息的范畴,但分割过程却能很容易将它们作为有效的动态目标像素分割出来.最近有关动态目标分割的研究提出了基于核心密度估计模型的分割方法.基于RGB颜色空间的核心密度估计模型能够在彩色视频中抑制阴影,但是对于灰度视频这种模型是无法抑制阴影的.作者针对上述局限提出了一种基于像素边缘测量的核心密度估计模型,用于动态目标分割,能够在彩色和灰度视频中抑制阴影.实验结果证明,这种新模型在通常的应用条件下是有效的.     

A survey of occlusion detection method for visual object

Occlusion problem is one of the challenging issues in vision field for a long time , and the occlu-sion phenomenon of visual object will be involved in many vision research fields .Once the occlusion occurs in a visual system , it will affect the effects of object recognition , tracking, observation and operation , so detecting occlusion autonomously should be one of the abilities for an intelligent vision system .The research on occlusion detection method for visual object has increasingly attracted atten-tions of scholars .First, the definition and classification of the occlusion problem are presented . Then, the characteristics and deficiencies of the occlusion detection methods based on the intensity image and the depth image are analyzed respectively , and the existing occlusion detection methods are compared.Finally, the problems of existing occlusion detection methods and possible research directions are pointed out .     

中国传统绘画中的六面体表现

与西方单点透视和埃及垂直投影等方法不同,中国绘画对六面体形象采取了平行后退的描绘方式。分析视觉生理及心理发现,由于视觉的形状恒常性及双眼视觉的机制,人在较近距离观察四边形时,会对其远离退缩的影像进行视觉矫正,因此中国画对六面体的独特描绘方式是具有生物学基础的,并由此构筑了独特的中国绘画空间表现方式。