Teaching an Old Robot New Tricks: Learning Novel Tasks via Interaction with People and Things

As AI has begun to reach out beyond its symbolic, objectivist roots into the embodied, experientialist realm, many projects are exploring different aspects of creating machines which interact with and respond to the world as humans do. Techniques for visual processing, object recognition, emotional response, gesture production and recognition, etc., are necessary components of a complete humanoid robot. However, most projects invariably concentrate on developing a few of these individual components, neglecting the issue of how all of these pieces would eventually fit together. The focus of the work in this dissertation is on creating a framework into which such specific competencies can be embedded, in a way that they can interact with each other and build layers of new functionality. To be of any practical value, such a framework must satisfy the real-world constraints of functioning in real-time with noisy sensors and actuators. The humanoid robot Cog provides an unapologetically adequate platform from which to take on such a challenge. This work makes three contributions to embodied AI. First, it offers a general-purpose architecture for developing behavior-based systems distributed over networks of PC's. Second, it provides a motor-control system that simulates several biological features which impact the development of motor behavior. Third, it develops a framework for a system which enables a robot to learn new behaviors via interacting with itself and the outside world. A few basic functional modules are built into this framework, enough to demonstrate the robot learning some very simple behaviors taught by a human trainer. A primary motivation for this project is the notion that it is practically impossible to build an "intelligent" machine unless it is designed partly to build itself. This work is a proof-of-concept of such an approach to integrating multiple perceptual and motor systems into a complete learning agent.

Model-Based Matching of Line Drawings by Linear Combinations of Prototypes

We describe a technique for finding pixelwise correspondences between two images by using models of objects of the same class to guide the search. The object models are 'learned' from example images (also called prototypes) of an object class. The models consist of a linear combination ofsprototypes. The flow fields giving pixelwise correspondences between a base prototype and each of the other prototypes must be given. A novel image of an object of the same class is matched to a model by minimizing an error between the novel image and the current guess for the closest modelsimage. Currently, the algorithm applies to line drawings of objects. An extension to real grey level images is discussed.

Global Depth Perception from Familiar Scene Structure

In the absence of cues for absolute depth measurements as binocular disparity, motion, or defocus, the absolute distance between the observer and a scene cannot be measured. The interpretation of shading, edges and junctions may provide a 3D model of the scene but it will not inform about the actual "size" of the space. One possible source of information for absolute depth estimation is the image size of known objects. However, this is computationally complex due to the difficulty of the object recognition process. Here we propose a source of information for absolute depth estimation that does not rely on specific objects: we introduce a procedure for absolute depth estimation based on the recognition of the whole scene. The shape of the space of the scene and the structures present in the scene are strongly related to the scale of observation. We demonstrate that, by recognizing the properties of the structures present in the image, we can infer the scale of the scene, and therefore its absolute mean depth. We illustrate the interest in computing the mean depth of the scene with application to scene recognition and object detection.

Stimulus Simplification and Object Representation: A Modeling Study

Tsunoda et al. (2001) recently studied the nature of object representation in monkey inferotemporal cortex using a combination of optical imaging and extracellular recordings. In particular, they examined IT neuron responses to complex natural objects and "simplified" versions thereof. In that study, in 42% of the cases, optical imaging revealed a decrease in the number of activation patches in IT as stimuli were "simplified". However, in 58% of the cases, "simplification" of the stimuli actually led to the appearance of additional activation patches in IT. Based on these results, the authors propose a scheme in which an object is represented by combinations of active and inactive columns coding for individual features. We examine the patterns of activation caused by the same stimuli as used by Tsunoda et al. in our model of object recognition in cortex (Riesenhuber 99). We find that object-tuned units can show a pattern of appearance and disappearance of features identical to the experiment. Thus, the data of Tsunoda et al. appear to be in quantitative agreement with a simple object-based representation in which an object's identity is coded by its similarities to reference objects. Moreover, the agreement of simulations and experiment suggests that the simplification procedure used by Tsunoda (2001) is not necessarily an accurate method to determine neuronal tuning.

Reconeixement de cd-roms musicals mitjançant visió per computador

Dissenyar, implementar i testejar un sistema per classificar imatges: disseny d’un sistema que primer aprèn com són les imatges d’una classe a partir d’un conjunt d’imatges d’entrenament i després és capaç de classificar noves imatges assignant-les-hi l’ etiqueta corresponent a una de les classes “apreses”. Concretament s’analitzen caràtules de cd-roms, les quals s’han de reconèixer per després reproduir automàticament la música del seu àlbum associat

Differential epipolar constraint in mobile robot egomotion estimation

The estimation of camera egomotion is a well established problem in computer vision. Many approaches have been proposed based on both the discrete and the differential epipolar constraint. The discrete case is mainly used in self-calibrated stereoscopic systems, whereas the differential case deals with a unique moving camera. The article surveys several methods for mobile robot egomotion estimation covering more than 0.5 million samples using synthetic data. Results from real data are also given

A concurrent region growing algorithm guided by circumscribed contours

Image segmentation of natural scenes constitutes a major problem in machine vision. This paper presents a new proposal for the image segmentation problem which has been based on the integration of edge and region information. This approach begins by detecting the main contours of the scene which are later used to guide a concurrent set of growing processes. A previous analysis of the seed pixels permits adjustment of the homogeneity criterion to the region's characteristics during the growing process. Since the high variability of regions representing outdoor scenes makes the classical homogeneity criteria useless, a new homogeneity criterion based on clustering analysis and convex hull construction is proposed. Experimental results have proven the reliability of the proposed approach

Estimating the motion of an underwater robot from a monocular image sequence

When underwater vehicles perform navigation close to the ocean floor, computer vision techniques can be applied to obtain quite accurate motion estimates. The most crucial step in the vision-based estimation of the vehicle motion consists on detecting matchings between image pairs. Here we propose the extensive use of texture analysis as a tool to ameliorate the correspondence problem in underwater images. Once a robust set of correspondences has been found, the three-dimensional motion of the vehicle can be computed with respect to the bed of the sea. Finally, motion estimates allow the construction of a map that could aid to the navigation of the robot

Surface registration from range image fusion

The registration of full 3-D models is an important task in computer vision. Range finders only reconstruct a partial view of the object. Many authors have proposed several techniques to register 3D surfaces from multiple views in which there are basically two aspects to consider. First, poor registration in which some sort of correspondences are established. Second, accurate registration in order to obtain a better solution. A survey of the most common techniques is presented and includes experimental results of some of them

Accuracy estimation of a new omnidirectional 3D vision sensor

We present a computer vision system that associates omnidirectional vision with structured light with the aim of obtaining depth information for a 360 degrees field of view. The approach proposed in this article combines an omnidirectional camera with a panoramic laser projector. The article shows how the sensor is modelled and its accuracy is proved by means of experimental results. The proposed sensor provides useful information for robot navigation applications, pipe inspection, 3D scene modelling etc

Robust decoupled visual servoing based on structured light

This paper focuses on the problem of realizing a plane-to-plane virtual link between a camera attached to the end-effector of a robot and a planar object. In order to do the system independent to the object surface appearance, a structured light emitter is linked to the camera so that 4 laser pointers are projected onto the object. In a previous paper we showed that such a system has good performance and nice characteristics like partial decoupling near the desired state and robustness against misalignment of the emitter and the camera (J. Pages et al., 2004). However, no analytical results concerning the global asymptotic stability of the system were obtained due to the high complexity of the visual features utilized. In this work we present a better set of visual features which improves the properties of the features in (J. Pages et al., 2004) and for which it is possible to prove the global asymptotic stability

Plane-to-plane positioning from image-based visual servoing and structured light

In this paper we face the problem of positioning a camera attached to the end-effector of a robotic manipulator so that it gets parallel to a planar object. Such problem has been treated for a long time in visual servoing. Our approach is based on linking to the camera several laser pointers so that its configuration is aimed to produce a suitable set of visual features. The aim of using structured light is not only for easing the image processing and to allow low-textured objects to be treated, but also for producing a control scheme with nice properties like decoupling, stability, well conditioning and good camera trajectory

Reconeixement d’escenes de la ciutat de Girona a partir del contingut d’imatges

Desenvolupament una aplicació informàtica basada en un sistema de visió per computador, la qual permeti donar una resposta en forma d'informació a partir d'una query d'una imatge que conté una escena o objecte en concret de manera que permeti reconèixer els objectes que apareixen en una imatge per llavors donar informació referent al contingut de la imatge a l’usuari que ha fet la consulta. Resumint, es tracta d’analitzar, dissenyar i construir un sistema de visió per computador capaç de reconèixer objectes d’interès en imatges

Classificació i reconeixement de vídeo

El principal objectiu d’aquest projecte és aconseguir classificar diferents vídeos d’esports segons la seva categoria. Els cercadors de text creen un vocabulari segons el significat de les diferents paraules per tal de poder identificar un document. En aquest projecte es va fer el mateix però mitjançant paraules visuals. Per exemple, es van intentar englobar com a una única paraula les diferents rodes que apareixien en els cotxes de rally. A partir de la freqüència amb què apareixien les paraules dels diferents grups dins d’una imatge vàrem crear histogrames de vocabulari que ens permetien tenir una descripció de la imatge. Per classificar un vídeo es van utilitzar els histogrames que descrivien els seus fotogrames. Com que cada histograma es podia considerar un vector de valors enters vàrem optar per utilitzar una màquina classificadora de vectors: una Support vector machine o SVM

Efficient 3D scene modeling and mosaicing

El modelat d'escenes és clau en un gran ventall d'aplicacions que van des de la generació mapes fins a la realitat augmentada. Aquesta tesis presenta una solució completa per a la creació de models 3D amb textura. En primer lloc es presenta un mètode de Structure from Motion seqüencial, a on el model 3D de l'entorn s'actualitza a mesura que s'adquireix nova informació visual. La proposta és més precisa i robusta que l'estat de l'art. També s'ha desenvolupat un mètode online, basat en visual bag-of-words, per a la detecció eficient de llaços. Essent una tècnica completament seqüencial i automàtica, permet la reducció de deriva, millorant la navegació i construcció de mapes. Per tal de construir mapes en àrees extenses, es proposa un algorisme de simplificació de models 3D, orientat a aplicacions online. L'eficiència de les propostes s'ha comparat amb altres mètodes utilitzant diversos conjunts de dades submarines i terrestres.