Landmark Objects Research Articles

A Novel Dataset For Intelligent Indoor Object Detection Systems

Indoor Scene understanding and indoor objects detection is a complex high-level task for automated systems applied to natural environments. Indeed, such a task requires huge annotated indoor images to train and test intelligent computer vision applications. One of the challenging questions is to adopt and to enhance technologies to assist indoor navigation for visually impaired people (VIP) and thus improve their daily life quality. This paper presents a new labeled indoor object dataset elaborated with a goal of indoor object detection (useful for indoor localization and navigation tasks). This dataset consists of 8000 indoor images containing 16 different indoor landmark objects and classes. The originality of the annotations comes from two new facts taken into account: (1) the spatial relationships between objects present in the scene and (2) actions possible to apply to those objects (relationships between VIP and an object).This collected dataset presents many specifications and strengths as it presents various data under various lighting conditions and complex image background to ensure more robustness when training and testing objects detectors. The proposed dataset, ready for use, provides 16 vital indoor object classes in order to contribute for indoor assistance navigation for VIP.

Learning Structural Representations via Dynamic Object Landmarks Discovery for Sketch Recognition and Retrieval.

State-of-the-art methods on sketch classification and retrieval are based on deep convolutional neural network to learn representations. Although deep neural networks have the ability to model images with hierarchical representations by convolution kernels, they can not automatically extract the structural representations of object categories in a human-perceptible way. Furthermore, sketch images usually have large scale visual variations caused by the styles of drawing or viewpoints, which make it difficult to develop generalized representations using the fixed computational mode of convolutional kernel. In this paper, our aim is to address the problem of fixed computational mode in feature extraction process without extra supervision. We propose a novel architecture to dynamically discover the object landmarks and learn the discriminative structural representations. Our model is composed of two components: a representative landmark discovering module that localizes the key points on the object, and a category-aware representation learning module that develops the category-specific features. Specifically, we develop a structure-aware offset layer to dynamically localize the representative landmarks, which is optimized based on the category labels without extra supervision. After that, a diversity branch is introduced to extract the global discriminative features for each category. Finally, we employ a multi-task loss function to develop an end-to-end trainable architecture. At testing time, we fuse all the predictions with different number of landmarks to achieve the final results. Through extensive experiments, we compare our model with several state-of-the-art methods on two challenging datasets TU-Berlin and Sketchy for sketch classification and retrieval, and the experimental results demonstrate the effectiveness of our proposed model.

Language cues in the formation of hierarchical representations of space

ABSTRACTWe study the role of place and region names in the formation of spatial hierarchies. Representations were probed behaviorally, i.e., by choices between two equidistant routes differing only in the number of regions they touched. Routes crossing fewer region boundaries are preferred over routes crossing more region boundaries. Visual landmark objects from a common semantic category placed in spatial vicinity induce region representations. Here we show that region dependent route preference is reduced if the landmark objects are replaced by signs displaying object names. The effects of explicit region names and linguistic properties of different naming schemes are discussed.

The contribution of visual attention and declining verbal memory abilities to age-related route learning deficits

Our ability to learn unfamiliar routes declines in typical and atypical ageing. The reasons for this decline, however, are not well understood. Here we used eye-tracking to investigate how ageing affects people’s ability to attend to navigationally relevant information and to select unique objects as landmarks. We created short routes through a virtual environment, each comprised of four intersections with two objects each, and we systematically manipulated the saliency and uniqueness of these objects. While salient objects might be easier to memorise than non-salient objects, they cannot be used as reliable landmarks if they appear more than once along the route. As cognitive ageing affects executive functions and control of attention, we hypothesised that the process of selecting navigationally relevant objects as landmarks might be affected as well. The behavioural data showed that younger participants outperformed the older participants and the eye-movement data revealed some systematic differences between age groups. Specifically, older adults spent less time looking at the unique, and therefore navigationally relevant, landmark objects. Both young and older participants, however, effectively directed gaze towards the unique and away from the non-unique objects, even if these were more salient. These findings highlight specific age-related differences in the control of attention that could contribute to declining route learning abilities in older age. Interestingly, route-learning performance in the older age group was more variable than in the young age group with some older adults showing performance similar to the young group. These individual differences in route learning performance were strongly associated with verbal and episodic memory abilities.

Possibilities of Involving the Semiotic Approach to the Urban Space Analysis

The main focus of the article is on the analysis of the heuristic potential of the semiotic approach in the study of urban space. Having identified the city as a cultural object as a research object, and starting from the methodological orientations of the sociocultural approach and post-structuralist analysis options, it is proposed to perceive the symbolic space of the city through the lens of semiotic optics as a subject area of research. The city as a socio-cultural phenomenon can therefore be considered as a collection of landmark objects. In order to understand the methodological potential of semiotics for city analysis, it is first of all necessary to consider the genesis of semiotic ideas and try to adapt them to city analysis as a landmark space. As a theoretical basis, the evolution of the views of a number of foreign scientists has been traced, in particular the attention has been paid to the scientific works of V. Benjamin, R. Bart, K. Levy-Strauss, C. Jenks, K. Lynch, and G. Simmel. The works of F. de Saussure, C. Pierce, C. Morris, Y. Lotman are distinguished as basic works in the field of semiotic analytics. An important aspect of the theoretical evolution of views on semiotic analysis is the proposed position of F. de Saussure on the basis of structuralism and poststructuralism, allowing the analysis of language, as well as any sign system, in the form of a structure with a fixed correspondence of signs and meanings. This thesis is present within the framework of structuralism and post structuralism. It is analyzed that the ideas of the "Frege triangle" and the "Oden-Richards triangle" in the study of cities significantly expand the potential of urban studies. It is emphasized that the use of tools of the semiotic approach allows to combine the analysis of the symbolic-symbolic system of the city and the space of urban everyday life.

Dissociable spatial memory systems revealed by typical and atypical human development.

Rodent lesion studies have revealed the existence of two causally dissociable spatial memory systems, localized to the hippocampus and striatum that are preferentially sensitive to environmental boundaries and landmark objects, respectively. Here we test whether these two memory systems are causally dissociable in humans by examining boundary- and landmark-based memory in typical and atypical development. Adults with Williams syndrome (WS)-a developmental disorder with known hippocampal abnormalities-and typical children and adults, performed a navigation task that involved learning locations relative to a boundary or a landmark object. We found that boundary-based memory was severely impaired in WS compared to typically-developing mental-age matched (MA) children and chronological-age matched (CA) adults, whereas landmark-based memory was similar in all groups. Furthermore, landmark-based memory matured earlier in typical development than boundary-based memory, consistent with the idea that the WS cognitive phenotype arises from developmental arrest of late maturing cognitive systems. Together, these findings provide causal and developmental evidence for dissociable spatial memory systems in humans.

Learning configural spatial knowledge from navigating in unstructured and unfamiliar virtual environments

Lynch’s (1960) description of the importance of districts for spatial memory was used as a basis for looking at the contributions of semantic meaning and physical structure to learning configural spatial knowledge about an unfamiliar virtual environment. According to Lynch, meaningful grouping, not just geometric structure should facilitate spatial learning, perhaps also assisted by structure. In a virtual environment consisting of four rooms/quadrants, both the presence or absence of walls and the semantic organization of the landmark objects (grouped versus distributed category members) were factorially combined. Both object locations and navigation paths were randomly determined. Angular error was reduced by the presence of walls and by the grouping of category members in rooms/quadrants, particularly when walls were present. Performance was better than chance even when object categories were distributed evenly throughout the environment and walls were not present. The theoretical and practical implications of these results were described.

IPhone otoscopes: Currently available, but reliable for tele-otoscopy in the hands of parents?

ObjectivesTele-otoscopy has been validated for tympanostomy surveillance and remote diagnosis when images are recorded by trained professionals. The CellScope iPhone Otoscope is a device that may be used for tele-otoscopy and it enables parents to record their children's ear examinations and send the films for remote physician diagnosis. This study aims to determine the ability to diagnose, and the reliability of the diagnosis when utilizing video exams obtained by a parent versus video exams obtained by an otolaryngologist. MethodsParents of children ages 17 years or younger attempted recordings of the tympanic membrane of their children with the CellScope after a video tutorial; a physician subsequently used the device to record the same ear. Recordings occurred prior to standard pediatric otolaryngology office evaluation. Later, a remote pediatric otolaryngologist attempted diagnosis solely based on the videos, blinded to whether the examination was filmed by a parent or physician. Interrater reliability between video diagnosis and original diagnosis on pneumatic otoscopy was measured, and objective tympanic membrane landmarks visualized on the films were recorded. ResultsEighty ears were enrolled and recorded. There was low interrater agreement (k = 0.42) between diagnosis based on parent videos as compared with pneumatic otoscopy. There was high agreement (k = 0.71) between diagnosis based on physician videos and pneumatic otoscopy. Physician videos and parent videos had only slight agreement on objective landmarks identified (k = 0.087). ConclusionsiPhone otoscopy provides reliable tele-otoscopy images in when used by trained professionals but, currently, images obtained by parents are not suitable for use in diagnosis.

Integration of objects and space in perception and memory.

Distinct processing of objects and space has been an organizing principle for studying higher-level vision and medial temporal lobe memory. Here, however, we discuss how object and spatial information are in fact closely integrated in vision and memory. The ventral, object-processing visual pathway carries precise spatial information, transformed from retinotopic coordinates into relative dimensions. At the final stages of the ventral pathway, including the dorsal anterior temporal lobe (TEd), object-sensitive neurons are intermixed with neurons that process large-scale environmental space. TEd projects primarily to perirhinal cortex (PRC), which in turn projects to lateral entorhinal cortex (LEC). PRC and LEC also combine object and spatial information. For example, PRC and LEC neurons exhibit place fields that are evoked by landmark objects or the remembered locations of objects. Thus, spatial information, on both local and global scales, is deeply integrated into the ventral (temporal) object-processing pathway in vision and memory.

Multi-Story Indoor Floor Plan Reconstruction via Mobile Crowdsensing

The lack of floor plans is a critical reason behind the current sporadic availability of indoor localization service. Service providers have to go through effort-intensive and time-consuming business negotiations with building operators, or hire dedicated personnel to gather such data. In this paper, we propose Jigsaw, a floor plan reconstruction system that leverages crowdsensed data from mobile users. It extracts the position, size, and orientation information of individual landmark objects from images taken by users. It also obtains the spatial relation between adjacent landmark objects from inertial sensor data, then computes the coordinates and orientations of these objects on an initial floor plan. By combining user mobility traces and locations where images are taken, it produces complete floor plans with hallway connectivity, room sizes, and shapes. It also identifies different types of connection areas (e.g., escalators and stairs) between stories, and employs a refinement algorithm to correct detection errors. Our experiments on three stories of two large shopping malls show that the 90-percentile errors of positions and orientations of landmark objects are about $1\sim 2m$ and $5\sim 9^\circ$ , while the hallway connectivity and connection areas between stories are 100 percent correct.

The Occipital Place Area Is Causally Involved in Representing Environmental Boundaries during Navigation

Thirty years of research suggests that environmentalboundaries-e.g., the walls of an experimental chamber or room-exert powerful influence on navigational behavior, often to the exclusion of other cues [1-9]. Consistent with this behavioral work, neurons in brain structures that instantiate spatial memory often exhibit firing fields that are strongly controlled by environmental boundaries [10-15]. Despite the clear importance of environmental boundaries for spatial coding, however, a brain region that mediates the perception of boundary information has not yet been identified. We hypothesized that the occipital place area (OPA), a scene-selective region located near the transverse occipital sulcus [16], might provide this perceptual source by extracting boundary information from visual scenes during navigation. To test this idea, we used transcranial magnetic stimulation (TMS) to interrupt processing in the OPA while subjects performed a virtual-reality memory task that required them to learn the spatial locations of test objects that were either fixed in place relative to the boundary of the environment or moved in tandem with a landmark object. Consistent with our prediction, we found that TMS to the right OPA impaired spatial memory for boundary-tethered, but not landmark-tethered, objects. Moreover, this effect was found when the boundary was defined by a wall, but not when it was defined by a marking on the ground. These results show that the OPA is causally involved in boundary-based spatial navigation and suggest that the OPA is the perceptual source of the boundary information that controls navigational behavior.

This Place Looks Familiar-How Navigators Distinguish Places with Ambiguous Landmark Objects When Learning Novel Routes.

We present two experiments investigating how navigators deal with ambiguous landmark information when learning unfamiliar routes. In the experiments we presented landmark objects repeatedly along a route, which allowed us to manipulate how informative single landmarks were (1) about the navigators' location along the route and (2) about the action navigators had to take at that location. Experiment 1 demonstrated that reducing location informativeness alone did not affect route learning performance. While reducing both location and action informativeness led to decreased route learning performance, participants still performed well above chance level. This demonstrates that they used other information than just the identity of landmark objects at their current position to disambiguate their location along the route. To investigate how navigators distinguish between visually identical intersections, we systematically manipulated the identity of landmark objects and the actions required at preceding intersections in Experiment 2. Results suggest that the direction of turn at the preceding intersections was sufficient to tell two otherwise identical intersections apart. Together, results from Experiments 1 and 2 suggest that route knowledge is more complex than simple stimulus-response associations and that neighboring places are tightly linked. These links not only encompass sequence information but also directional information which is used to identify the correct direction of travel at subsequent locations, but can also be used for self-localization.

The hippocampus

The hippocampus

The Occipital Place Area is causally involved in representing environmental boundaries during navigation

Thirty years of research suggests that environmental boundaries-e.g., the walls of an experimental chamber or room-exert powerful influence on navigational behavior, often to the exclusion of other cues [1-9]. Consistent with this behavioral work, neurons in brain structures that instantiate spatial memory often exhibit firing fields that are strongly controlled by environmental boundaries [10-15]. Despite the clear importance of environmental boundaries for spatial coding, however, a brain region that mediates the perception of boundary information has not yet been identified. We hypothesized that the occipital place area (OPA), a scene-selective region located near the transverse occipital sulcus [16], might provide this perceptual source by extracting boundary information from visual scenes during navigation. To test this idea, we used transcranial magnetic stimulation (TMS) to interrupt processing in the OPA while subjects performed a virtual-reality memory task that required them to learn the spatial locations of test objects that were either fixed in place relative to the boundary of the environment or moved in tandem with a landmark object. Consistent with our prediction, we found that TMS to the right OPA impaired spatial memory for boundary-tethered, but not landmark-tethered, objects. Moreover, this effect was found when the boundary was defined by a wall, but not when it was defined by a marking on the ground. These results show that the OPA is causally involved in boundary-based spatial navigation and suggest that the OPA is the perceptual source of the boundary information that controls navigational behavior.

How to construct a linguistic landmark: language cues in the formation of hierarchical representations of space.

The formation of a hierarchical representation of space can be induced by the spatial adjacency of places marked by landmark objects from the same semantic category, as was demonstrated by a route-planning experiment (Wiener and Mallot in Spat Cogn Comput 3(4):331-358, 2003). Using the same paradigm, we tested the efficiency of linguistic cues with various hierarchical categorization principles in regional structuring. In five different conditions, places of the experimental environment were characterized (1) by landmark objects, (2) by arbitrary proper names on region level, (3) by prototypical names on region level, (4) by nouns from different semantic categories, and (5) by groups of nouns with intrinsic whole-part relation. A region effect comparable to the landmark object condition was found only for the whole-part relations, which combine spatial proximity with a shared context. A linguistic analysis revealed a correspondence to the intended regional subdivision only for the cues of this condition.

Star formation inz> 1 3CR host galaxies as seen byHerschel

We present Herschel (PACS and SPIRE) far-infrared (FIR) photometry of a complete sample of z>1 3CR sources, from the Herschel GT project The Herschel Legacy of distant radio-loud AGN (PI: Barthel). Combining these with existing Spitzer photometric data, we perform an infrared (IR) spectral energy distribution (SED) analysis of these landmark objects in extragalactic research to study the star formation in the hosts of some of the brightest active galactic nuclei (AGN) known at any epoch. Accounting for the contribution from an AGN-powered warm dust component to the IR SED, about 40% of our objects undergo episodes of prodigious, ULIRG-strength star formation, with rates of hundreds of solar masses per year, coeval with the growth of the central supermassive black hole. Median SEDs imply that the quasar and radio galaxy hosts have similar FIR properties, in agreement with the orientation-based unification for radio-loud AGN. The star-forming properties of the AGN hosts are similar to those of the general population of equally massive non-AGN galaxies at comparable redshifts, thus there is no strong evidence of universal quenching of star formation (negative feedback) within this sample. Massive galaxies at high redshift may be forming stars prodigiously, regardless of whether their supermassive black holes are accreting or not.

Visual landmark recognition from Internet photo collections: A large-scale evaluation

The task of a visual landmark recognition system is to identify photographed buildings or objects in query photos and to provide the user with relevant information on them. With their increasing coverage of the world’s landmark buildings and objects, Internet photo collections are now being used as a source for building such systems in a fully automatic fashion. This process typically consists of three steps: clustering large amounts of images by the objects they depict; determining object names from user-provided tags; and building a robust, compact, and efficient recognition index. To this date, however, there is little empirical information on how well current approaches for those steps perform in a large-scale open-set mining and recognition task. Furthermore, there is little empirical information on how recognition performance varies for different types of landmark objects and where there is still potential for improvement. With this paper, we intend to fill these gaps. Using a dataset of 500k images from Paris, we analyze each component of the landmark recognition pipeline in order to answer the following questions: How many and what kinds of objects can be discovered automatically? How can we best use the resulting image clusters to recognize the object in a query? How can the object be efficiently represented in memory for recognition? How reliably can semantic information be extracted? And finally: What are the limiting factors in the resulting pipeline from query to semantics? We evaluate how different choices of methods and parameters for the individual pipeline steps affect overall system performance and examine their effects for different query categories such as buildings, paintings or sculptures.

Depth Sensor Based Automatic Hand Region Extraction by Using Time-series Curve and Its Application to Japanese Finger-spelled Sign Language Recognition

Abstract Hand sign recognition is one of most challenging issues in computer vision and human computer interaction, and many researchers tackle this issue. In this research, we focus on JFSL (Japanese Finger-spelled Sign Language) which is one of hand signs. The tasks for achieving high performance of JFSL recognition as well as other hand signs are how to extract hand region precisely and how to recognize hand signs accurately. To deal with the former task, in this paper, we propose an automatic hand region extraction method with a depth sensor. The characteristic points of our proposed method are to utilize Time-Series Curve, which is one of contour features, and to extract hand region accurately without wearing landmark object such as a color wristband. On the other hand, to tackle the latter task, in this research, we focus on a deep neural network based recognition method since such a method is reported that it allows us to achieve high performance for various recognition tasks. Therefore, in this paper, we investigate JFSL recognition performance with a deep neural network approach compared to that with the conventional image recognition method (HOG+SVM). From the experimental results with 8 subjects, we have confirmed that our proposed method allows us to extract hand region accurately regardless of subjects and JFSL signs. In addition, from the experimental results with a deep neural network based recognition method for JFSL recognition, we have achieved at least average recognition rate over 88%.

Fine-tuning regression forests votes for object alignment in the wild.

In this paper, we propose a object alignment method that detects the landmarks of an object in 2D images. In the regression forests (RFs) framework, observations (patches) that are extracted at several image locations cast votes for the localization of several landmarks. We propose to refine the votes before accumulating them into the Hough space, by sieving and/or aggregating. In order to filter out false positive votes, we pass them through several sieves, each associated with a discrete or continuous latent variable. The sieves filter out votes that are not consistent with the latent variable in question, something that implicitly enforces global constraints. In order to aggregate the votes when necessary, we adjusts on-the-fly a proximity threshold by applying a classifier on middle-level features extracted from voting maps for the object landmark in question. Moreover, our method is able to predict the unreliability of an individual object landmark. This information can be useful for subsequent object analysis like object recognition. Our contributions are validated for two object alignment tasks, face alignment and car alignment, on data sets with challenging images collected in the wild, i.e., the Labeled Face in the Wild, the Annotated Facial Landmarks in the Wild, and the street scene car data set. We show that with the proposed approach, and without explicitly introducing shape models, we obtain performance superior or close to the state of the art for both tasks.

Interpretation of plural definites in discourse

In this paper we offer a study on the interpretation of plural definites in discourse (‘the tank engines’) and their interaction with spatial adpositions (‘to’ and ‘at’). The novel empirical findings in the paper support the following assumptions on the contribution of spatial adpositions to the interpretation of plural definites. First, the interpretation of plural definites can be influenced by the lexical aspect type of adpositions. While ‘to’ as ‘telic’ predicate can license both a ‘collective’ and a ‘distributive’ reading for plural definites, ‘at’ as an ‘atelic’ predicate only licenses a ‘collective’ reading. Second, the precise lexical content of adpositions determines which interpretation is accessed. It is claimed that ‘at’ denotes a ‘general location’ relation between locatum and landmark object, and thus licenses a collective reading for plural definites.