Ana C. Murillo

SURF features for efficient robot localization with omnidirectional images

Many robotic applications work with visual reference maps, which usually consist of sets of more or less organized images. In these applications, there is a compromise between the density of reference data stored and the capacity to identify later the robot localization, when it is not exactly in the same position as one of the reference views. Here we propose the use of a recently developed feature, SURF, to improve the performance of appearance-based localization methods that perform image retrieval in large data sets. This feature is integrated with a vision-based algorithm that allows both topological and metric localization using omnidirectional images in a hierarchical approach. It uses pyramidal kernels for the topological localization and three-view geometric constraints for the metric one. Experiments with several omnidirectional images sets are shown, including comparisons with other typically used features (radial lines and SIFT). The advantages of this approach are proved, showing the use of SURF as the best compromise between efficiency and accuracy in the results.

Visual door detection integrating appearance and shape cues

An important component of human-robot interaction is the capability to associate semantic concepts with encountered locations and objects. This functionality is essential for visually guided navigation as well as location and object recognition. In this paper we focus on the problem of door detection using visual information only. Doors are frequently encountered in structured man-made environments and function as transitions between different places. We adopt a probabilistic approach for door detection, by defining the likelihood of various features for generated door hypotheses. Differing from previous approaches, the proposed model captures both the shape and appearance of the door. This is learned from a few training examples, exploiting additional assumptions about the structure of indoor environments. After the learning stage, we describe a hypothesis generation process and several approaches to evaluate the likelihood of the generated hypotheses. The approach is tested on numerous examples of indoor environment. It shows a good performance provided that the door extent in the images is sufficiently large and well supported by low level feature measurements.

From omnidirectional images to hierarchical localization

We propose a new vision-based method for global robot localization using an omnidirectional camera. Topological and metric localization information are combined in an efficient, hierarchical process, with each step being more complex and accurate than the previous one but evaluating fewer images. This allows us to work with large reference image sets in a reasonable amount of time. Simultaneously, thanks to the use of 1D three-view geometry, accurate metric localization can be achieved based on just a small number of nearby reference images. Owing to the wide baseline features used, the method deals well with illumination changes and occlusions, while keeping the computational load small. The simplicity of the radial line features used speeds up the process without affecting the accuracy too much. We show experiments with two omnidirectional image data sets to evaluate the performance of the method and compare the results using the proposed radial lines with results from state-of-the-art wide-baseline matching techniques.

Localization in Urban Environments Using a Panoramic Gist Descriptor

Vision-based topological localization and mapping for autonomous robotic systems have received increased research interest in recent years. The need to map larger environments requires models at different levels of abstraction and additional abilities to deal with large amounts of data efficiently. Most successful approaches for appearance-based localization and mapping with large datasets typically represent locations using local image features. We study the feasibility of performing these tasks in urban environments using global descriptors instead and taking advantage of the increasingly common panoramic datasets. This paper describes how to represent a panorama using the global gist descriptor, while maintaining desirable invariance properties for location recognition and loop detection. We propose different gist similarity measures and algorithms for appearance-based localization and an online loop-closure detection method, where the probability of loop closure is determined in a Bayesian filtering framework using the proposed image representation. The extensive experimental validation in this paper shows that their performance in urban environments is comparable with local-feature-based approaches when using wide field-of-view images.

Localization and Matching Using the Planar Trifocal Tensor With Bearing-Only Data

This paper addresses the robot and landmark localization problem from bearing-only data in three views, simultaneously to the robust association of this data. The localization algorithm is based on the 1-D trifocal tensor, which relates linearly the observed data and the robot localization parameters. The aim of this work is to bring this useful geometric construction from computer vision closer to robotic applications. One contribution is the evaluation of two linear approaches of estimating the 1-D tensor: the commonly used approach that needs seven bearing-only correspondences and another one that uses only five correspondences plus two calibration constraints. The results in this paper show that the inclusion of these constraints provides a simpler and faster solution and better estimation of robot and landmark locations in the presence of noise. Moreover, a new method that makes use of scene planes and requires only four correspondences is presented. This proposal improves the performance of the two previously mentioned methods in typical man-made scenarios with dominant planes, while it gives similar results in other cases. The three methods are evaluated with simulation tests as well as with experiments that perform automatic real data matching in conventional and omnidirectional images. The results show sufficient accuracy and stability to be used in robotic tasks such as navigation, global localization or initialization of simultaneous localization and mapping (SLAM) algorithms.

Urban tribes: Analyzing group photos from a social perspective

The explosive growth in image sharing via social networks has produced exciting opportunities for the computer vision community in areas including face, text, product and scene recognition. In this work we turn our attention to group photos of people and ask the question: what can we determine about the social subculture or urban tribe to which these people belong? To this end, we propose a framework employing low- and mid-level features to capture the visual attributes distinctive to a variety of urban tribes. We proceed in a semi-supervised manner, employing a metric that allows us to extrapolate from a small number of pairwise image similarities to induce a set of groups that visually correspond to familiar urban tribes such as biker, hipster or goth. Automatic recognition of such information in group photos offers the potential to improve recommendation services, context sensitive advertising and other social analysis applications. We present promising preliminary experimental results that demonstrate our ability to categorize group photos in a socially meaningful manner.

Experiments in place recognition using gist panoramas

In this paper we investigate large scale view based localization in urban areas using panoramic images. The presented approach utilizes global gist descriptor computed for portions of panoramic images and a simple similarity measure between two panoramas, which is robust to changes in vehicle orientation, while traversing the same areas in different directions. The global gist feature [14] has been demonstrated previously to be a very effective conventional image descriptor, capturing the basic structure of different types of scenes in a very compact way. We present an extensive experimental validation of our panoramic gist approach on a large scale Street View data set of panoramic images for place recognition or topological localization.

Localization with omnidirectional images using the radial trifocal tensor

In this paper we present a technique to linearly recover 2D structure and motion in man made environments from three uncalibrated omnidirectional views. We use vertical lines from the scene which are projected as radial lines in the images and are automatically matched. The algorithm is based on a 1D radial trifocal tensor which encodes the relations of the three views and the projected lines. We include experiments with real images, which demonstrate the good performance of the method and its application to robotic tasks, such as robot localization based in a database of reference images or to obtain the initial values of robot and landmarks localization in SLAM algorithms

Semantic labeling for indoor topological mapping using a wearable catadioptric system

An important part of current research on appearance based mapping goes towards richer semantic representations of the environment, which may allow autonomous systems to perform higher level tasks and provide better human-robot interaction. This work presents a new omnidirectional vision based scene labeling approach for augmented indoor topological mapping. Omnidirectional vision systems are of particular interest because they allow us to have more compact and efficient representation of the environment. Our proposal includes novel ideas in order to augment the semantic information of a typical indoor topological map: we pay special attention to the semantic labels of the different types of transitions between places, and propose a simple way to include this semantic information to build a topological map, as part of the criteria to segment the environment. This work is built on efficient catadioptric image representation based on the Gist descriptor, which is used to classify the acquired views into types of indoor regions. The basic types of indoor regions considered are Place and Transition, farthest divided into more specific subclasses, e.g., Transition into door, stairs and elevator. Besides using the result of this labeling, the proposed mapping approach includes a probabilistic model to account for spatio-temporal consistency. All the proposed ideas have been evaluated in a new indoor dataset presented in this paper. This dataset has been acquired with our wearable catadioptric vision system 1 , showing promising results

From Bikers to Surfers: Visual Recognition of Urban Tribes.

Iljung S. Kwak1 iskwak@cs.ucsd.edu Ana C. Murillo2 acm@unizar.es Peter N. Belhumeur3 belhumeur@cs.columbia.edu David Kriegman1 kriegman@cs.ucsd.edu Serge Belongie1 sjb@cs.ucsd.edu 1 Dept. of Computer Science and Engineering University of California, San Diego, USA. 2 Dpt. Informatica e Ing. Sistemas Inst. Investigacion en Ingenieria de Aragon. University of Zaragoza, Spain. 3 Department of Computer Science Columbia University, USA.