Unified framework for recognition, localization and mapping using wearable cameras

Monocular approaches to simultaneous localization and mapping (SLAM) have recently addressed with success the challenging problem of the fast computation of dense reconstructions from a single, moving camera. Thus, if these approaches initially relied on the detection of a reduced set of interest points to estimate the camera position and the map, they are currently able to reconstruct dense maps from a handheld camera while the camera coordinates are simultaneously computed. However, these maps of 3-dimensional points usually remain meaningless, that is, with no memorable items and without providing a way of encoding spatial relationships between objects and paths. In humans and mobile robotics, landmarks play a key role in the internalization of a spatial representation of an environment. They are memorable cues that can serve to define a region of the space or the location of other objects. In a topological representation of the space, landmarks can be identified and located according to its structural, perceptive or semantic significance and distinctiveness. But on the other hand, landmarks may be difficult to be located in a metric representation of the space. Restricted to the domain of visual landmarks, this work describes an approach where the map resulting from a point-based, monocular SLAM is annotated with the semantic information provided by a set of distinguished landmarks. Both features are obtained from the image. Hence, they can be linked by associating to each landmark all those point-based features that are superimposed to the landmark in a given image (key-frame). Visual landmarks will be obtained by means of an object-based, bottom-up attention mechanism, which will extract from the image a set of proto-objects. These proto-objects could not be always associated with natural objects, but they will typically constitute significant parts of these scene objects and can be appropriately annotated with semantic information. Moreover, they will be affine covariant regions, that is, they will be invariant to affine transformation, being detected under different viewing conditions (view-point angle, rotation, scale, etc.). Monocular SLAM will be solved using the accurate parallel tracking and mapping (PTAM) framework by Klein and Murray in Proceedings of IEEE/ACM international symposium on mixed and augmented reality, 2007.     

Perceptual organization and artificial attention for visual landmarks detection

In biological vision systems, attention mechanisms are responsible for selecting the relevant information from the sensed field of view, so that the complete scene can be analyzed using a sequence of rapid eye saccades. In recent years, efforts have been made to imitate such attention behavior in artificial vision systems, because it allows optimizing the computational resources as they can be focused on the processing of a set of selected regions. In the framework of mobile robotics navigation, this work proposes an artificial model where attention is deployed at the level of objects (visual landmarks) and where new processes for estimating bottom-up and top-down (target-based) saliency maps are employed. Bottom-up attention is implemented through a hierarchical process, whose final result is the perceptual grouping of the image content. The hierarchical grouping is applied using a Combinatorial Pyramid that represents each level of the hierarchy by a combinatorial map. The process takes into account both image regions (faces in the map) and edges (arcs in the map). Top-down attention searches for previously detected landmarks, enabling their re-detection when the robot presumes that it is revisiting a known location. Landmarks are described by a combinatorial submap; thus, this search is conducted through an error-tolerant submap isomorphism procedure.     

A saliency map in primary visual cortex

I propose that pre-attentive computational mechanisms in primary visual cortex create a saliency map. This map awards higher responses to more salient image locations; these responses are those of conventional V1 cells tuned to input features, such as orientation and color. Hence no separate feature maps, or any subsequent combination of them, is needed to create a saliency map. I use a model to show that this saliency map accounts for the way that the relative difficulties of visual search tasks depend on the features and spatial configurations of targets and distractors. This proposal links psychophysical behavior to V1 physiology and anatomy, and thereby makes testable predictions.     

幼儿地图定向任务中几何线索的运用及发展

几何线索对儿童理解地图与现实空间之间的对应关系具重要意义。目前研究以几何地图为材料,通过定向任务,考察儿童对不同几何线索(拓扑关系、欧式几何特征、度量结构)的利用与发展,揭示了儿童对地图与现实空间之间对应关系的认知过程。研究发现,3岁儿童已初步发展出使用欧氏几何特征和度量结构表征空间位置的能力;但不同欧式几何特征(相对距离、角度、方向)被儿童掌握的时间存在明显差别;关于拓扑关系的研究较少,尚未掌握儿童利用该线索的发展规律。提高生态效度是未来研究需解决的主要问题。     

Skill transfer through goal-driven representation mapping

In this paper, we present an approach to transfer that involves analogical mapping of symbols across different domains. We relate this mechanism to Icarus, a theory of the human cognitive architecture. Our system can transfer skills across domains hypothesizing maps between representations, improving performance in novel domains. Unlike previous approaches to analogical transfer, our method uses an explanatory analysis that compares how well a new domain theory explains previous solutions under different mapping hypotheses. We present experimental evidence that the new mechanism improves transfer over Icarus’ basic learning processes. Moreover, we argue that the same features which distinguish Icarus from other architectures support representation mapping in a natural way and operate synergistically with it. These features enable our analogy system to translate a map among concepts into a map between skills, and to support transfer even if two domains are only partially analogous. We also discuss our system’s relation to other work on analogy and outline directions for future research.     

识图方式与地图难度对定向运动员识图决策绩效与视觉搜索特征的影响

目的：研究旨在探究定向运动员在识别定向运动地图时的决策绩效与视觉搜索特征。方法：研究采用组内设计,操控地图难度（简单和复杂）和运动员识图方式（精确识图和概略识图）对高水平定向运动员进行实验研究。结果：（1）随着地图难度的加大,定向运动员识图准确性下降,概略识图反应时增加。（2）识图时,简单地图比复杂地图注视频率和眼跳距离更大,注视次数更少;概略识图比精确识图注视次数更多、注视频率更高、眼跳距离更大;概略识图时,简单地图条件下运动员从起点开始正向搜索,复杂地图条件下从终点开始逆向搜索,注视区域面积大且分散;精确识图时,无论是简单地图还是复杂地图,所有运动员均首先搜索检查点说明表和终点,注视区域面积小且集中。结论：地图难度制约着定向运动员的识图决策绩效。定向运动员识图的视觉搜索特征受识图方式与地图难度影响,表现出不同的视觉搜索策略。     

Robots that imitate humans

The study of social learning in robotics has been motivated by both scientific interest in the learning process and practical desires to produce machines that are useful, flexible, and easy to use. In this review, we introduce the social and task-oriented aspects of robot imitation. We focus on methodologies for addressing two fundamental problems. First, how does the robot know what to imitate? And second, how does the robot map that perception onto its own action repertoire to replicate it? In the future, programming humanoid robots to perform new tasks might be as simple as showing them.     

Image representation,scaling and cognitive model of object perception

Neuropsychological investigations of visual imagery and representations have led to a deeper understanding of the spatial perception, representation and memory. But how each individual perceives object’s geometrical properties and how they differ from person to person, both under event-related memory and normal recollecting memory in the presence or in the absence of direct sensory stimulation is still unclear. Spatial knowledge is diverse, complex, and multi-modal, as are the situations in which it is used. All seem to agree that a cognitive map is a mental representation of an external environment. The image scaling is important in understanding the psychological dysfunctions of patients suffering from spatial cognition problems. The scaling becomes self-evident in art forms, when people are asked to draw image of objects they see actively or from their short or long term memory. In this paper we develop a comprehensive model of this scaling factor and its implications in spatial image representation and memory. We also extend its notion in understanding the perception of objects whose representations are normally not possible (like the perception of universal scales, infinities and parallel lines) but are well comprehended by the human brains. Here we give a scaling factor which is variable depending on the situations for a person based on his visual memory and drawing capabilities. And then extend it to analyse his cognitive strengths, disorders and any imperfections. This model also helps in formalizing the architectural cognitive maps needed to change the scaling factor, depending on the types of visual works one performs.     

A continuous smooth map of space in the primary visual cortex of the common marmoset

We examined the fine-scale mapping of the visual world within the primary visual cortex of the marmoset monkey (Callithrix jacchus) using differential optical imaging. We stimulated two sets of complementary stripe-like locations in turn, subtracting them to generate the cortical representations of continuous bands of visual space. Rotating this stimulus configuration makes it possible to map different spatial axes within the primary visual cortex. In a similar manner, shifting the stimulated locations between trials makes it possible to map retinotopy at an even finer scale. Using these methods we found no evidence of any local anisotropies or distortions in the cortical representation of visual space. This is despite the fact that orientation preference is mapped in a discontinuous manner across the surface of marmoset V1. Overall, our results indicate that space is mapped in a continuous and smooth manner in the primary visual cortex of the common marmoset.     

听觉诱发表象的眼动机制

视觉心理表象是一种基于过去经验的自上而下的加工过程,它是信息表征的一种形式.以往的研究主要采用视觉来诱发表象,但是这很难排除视觉后像的干扰,而且与表象“自上而下”的加工过程不一致.听觉诱发表象很好地解决了视觉后像的干扰问题,也更吻合表象“自上而下”的加工过程,因为被试产生的表象是基于自己过去的经验,从自身出发,主动加工的结果.采用实验法,考察通过听觉通道诱发表象眼动的情况,使用眼动仪记录被试的眼动轨迹.结果发现：（1）不管是运动还是静止的物体,都能诱发表象,在表象时存在眼动行为.（2）运动的物体诱发眼动的程度更大.（3）眼动影响表象的建构,在表象建构中起机能性作用.     

A single sensor system for mapping in GNSS-denied environments

Autonomous robotic vehicles need accurate positioning to navigate. For outdoor autonomous vehicles, the localization problem has been solved using GNSS systems. However, many places suffer from problems in the signal of those systems, known as GNSS-denied environments. To face such a problem, several approaches first map the environment to thereafter localize the vehicle within it. The solutions for mapping in GNSS-denied places, mostly, are tested indoors, and the revisited areas are generally close from the starting position. In this work, we develop a single sensor system for mapping in large-scale GNSS-denied sites, based exclusively on a 3D LiDAR system. The proposed work consists of a 3D-LiDAR with a LiDAR Odometry approach (LO) estimating movements between frames thus providing of dead-reckoning estimates of the vehicle. The LiDAR Odometry is the input to a virtual GNSS system based on a Particle Filter Localization which matches the estimated dead-reckoning with a road map, with the assumption that the vehicle usually navigates on roads. The global position produced by the virtual GNSS is used to detect and correct loop closures in case of revisited areas. A GraphSLAM implementation fuses the outcomes from LiDAR Odometry, virtual GNSS and loop closure, and yield a feasible pose of the vehicle. Finally, a mapping procedure places every 3D frame and builds 2D occupancy grid maps. The system developed here is evaluated empirically using two datasets collected in a dynamic environment with paths of 3.7 and 6.5 km long, respectively. For both datasets, the system presents an RMS of 6.5 m according to GNSS sensor readings used for comparison purposes.     

CONCEPT MAPPING AS A FEMINIST RESEARCH METHOD

A method used in program evaluation and public health research called concept mapping is examined in this article for its usefulness in feminist research. This method embodies several defining characteristics of feminist social science. Concept mapping is a single method that integrates qualitative and quantitative approaches, provides an opportunity for participants to work together as a group to develop an understanding of a concept, and places the participants in control of interpretation. Over the course of a six-step process, a group of people are assembled to discuss an issue or concept, moving toward a group understanding of that concept, which is then represented in a visual picture, or map. Concept mapping was used in the current study to examine how community resources for sexual-assault victims could be improved. A national random sample of 168 rape-victim advocates provided ideas as to how the legal, medical, and mental health systems could better serve victims. A subgroup of advocates then constructed and interpreted a concept map. The map suggested that rape victims still face many problems in seeking community help. Twelve clusters of broad-based and specific system changes were identified (e.g., fighting victim blaming, community education, sensitizing medical staff, legal reform). Implications for research on sexual assault and feminist research methodology are discussed.     

Early development of scaling ability

The map is a small-scaled version of the space it represents. It has been argued that children have difficulty interpreting maps because they do not understand scale relations. Recent research has shown that even preschoolers can solve problems that involve scaling in one dimension. This study examined whether early scaling ability extends to tasks involving two-dimensional maps and referent spaces of different sizes. Results showed that about 60% of the 4-year-olds and 90% of the 5-year-olds tested used distance information presented on a map to locate an object in a two-dimensional spatial layout. Children had more difficulties in solving mapping tasks with a larger referent space. This decrease in accuracy as a function of space size on the mapping task was greater than would have been expected on the basis of performance on a parallel nonmapping task. The results are discussed in terms of their implications for the mechanisms underlying early scaling ability.     

On Berliner's Thinking in Jazz: The Infinite Are of Improvisation

In order to characterize divergent thinking, the author performed self-organization feature mapping on the multidimensional data of Williams's (1980) Creative Thinking Tests. The resultant population map of cluster was obtained. Principal-components analysis was also applied to determine the most significant features characterizing divergent thinking. Using the two most significant features to define a 2-dimensional space, the author ascertained characterizations of each cluster over the population map. The results obtained compared favorably with the topological structure of the population map. A secondary differentiation was found in the divergent thinking test. The topological structure for divergent thinking should prove useful for characterization of creative thinking.     

Thinking graphically: Connecting vision and cognition during graph comprehension

Task analytic theories of graph comprehension account for the perceptual and conceptual processes required to extract specific information from graphs. Comparatively, the processes underlying information integration have received less attention. We propose a new framework for information integration that highlights visual integration and cognitive integration. During visual integration, pattern recognition processes are used to form visual clusters of information; these visual clusters are then used to reason about the graph during cognitive integration. In 3 experiments, the processes required to extract specific information and to integrate information were examined by collecting verbal protocol and eye movement data. Results supported the task analytic theories for specific information extraction and the processes of visual and cognitive integration for integrative questions. Further, the integrative processes scaled up as graph complexity increased, highlighting the importance of these processes for integration in more complex graphs. Finally, based on this framework, design principles to improve both visual and cognitive integration are described.     

The equivalent background of bleaching

Stiles and Crawford proposed that a retinal region bleached by preexposure to intense light behaves as if it were illuminated by some steady veiling or background luminance. We test this notion by comparing the afterimage of a bleaching light with a steady (and retinally stabilized) light of adjustable intensity, in the manner of Barlow and Sparrock. With their matching procedure, and also with a new procedure, we find as they did that during the rod phase of recovery the afterimage does look like a stabilized field of an intensity which, presented as a background, brings visual sensitivity to the same level. It is as if the two conditions produce equal signals at some stage of the visual pathway. Liked Barlow and Sparrock we observe a rod-cone break in the afterimage matches. However, we argue that the appearance of the rod-cone break presents a paradox and we show a way to resolve it.     

Functional equivalence of spatial images from touch and vision: evidence from spatial updating in blind and sighted individuals

This research examined whether visual and haptic map learning yield functionally equivalent spatial images in working memory, as evidenced by similar encoding bias and updating performance. In 3 experiments, participants learned 4-point routes either by seeing or feeling the maps. At test, blindfolded participants made spatial judgments about the maps from imagined perspectives that were either aligned or misaligned with the maps as represented in working memory. Results from Experiments 1 and 2 revealed a highly similar pattern of latencies and errors between visual and haptic conditions. These findings extend the well-known alignment biases for visual map learning to haptic map learning, provide further evidence of haptic updating, and most important, show that learning from the 2 modalities yields very similar performance across all conditions. Experiment 3 found the same encoding biases and updating performance with blind individuals, demonstrating that functional equivalence cannot be due to visual recoding and is consistent with an amodal hypothesis of spatial images.