Frank P. Ferrie

Autonomous exploration: driven by uncertainty

Passively accepting measurements of the world is not enough, as the data we obtain is always incomplete, and the inferences made from it uncertain to a degree which is often unacceptable. If we are to build machines that operate autonomously, they will always be faced with this dilemma, and can only be successful if they play a much more active role. This paper presents such a machine. It deliberately seeks out those parts of the world which maximize the fidelity of its internal representations, and keeps searching until those representations are acceptable. We call this paradigm autonomous exploration, and the machine an autonomous explorer. This paper has two major contributions. The first is a theory that tells us how to explore, and which confirms the intuitive ideas we have put forward previously. The second is an implementation of that theory. In our laboratory, we have constructed a working autonomous explorer and here, for the first time, show it in action. The system is entirely bottom-up and does not depend on any a priori knowledge of the environment. To our knowledge, it is the first to have successfully closed the loop between gaze planning and the inference of complex 3D models.

Darboux frames, snakes, and super-quadrics: geometry from the bottom up

A representational and a computational model for deriving 3-D articulated volumetric descriptions of objects from laser rangefinder data is described. This method is purely bottom up: it relies on general assumptions cast in terms of differential geometry. Darboux frames, snakes, and superquadrics form the basis of this representation, and curvature consistency provides the computational framework. The organization is hierarchical. Darboux frames are used to describe the local surface, whereas snakes are used to interpolate between features, particularly those that serve to partition a surface into its constituent parts. Superquadrics are used to characterize the 3-D shape of each surface partition. The result is a set of connected volumetric primitives that serve to describe the overall shape of an object. Examples that show how the approach performs on data acquired with a laser rangefinder are included. >

Viewpoint selection by navigation through entropy maps

In this paper, we show how entropy maps can be used to guide an active observer along an optimal trajectory, by which the identity and pose of objects in the world can be inferred with confidence, while minimizing the amount of data that must be gathered. Specifically we consider the case of active object recognition where entropy maps are used to encode prior knowledge about the discriminability of objects as a function of viewing position. The paper describes how these maps are computed using optical flow signatures as a case study, and how a gaze-planning strategy can be formulated by using entropy minimization as a basis for choosing a next best view. Experimental results are presented which show the strategys effectiveness for active object recognition using a single monochrome television camera.

An integrated range-sensing, segmentation and registration framework for the characterization of intra-surgical brain deformations in image-guided surgery

Image-guided surgery (IGS) is a technique for localizing anatomical structures on the basis of volumetric image data and for determining the optimal surgical path to reach these structures, by the means of a localization device, or probe, whose position is tracked over time. The usefulness of this technology hinges on the accuracy of the transformation between the image volume and the space occupied by the patient anatomy and spanned by the probe. Unfortunately, in neurosurgery this transformation can be degraded by intra-surgical brain shift, which often measures more than 10 mm and can exceed 25 mm. We propose a method for characterizing brain shift that is based on non-rigid surface registration, and can be combined with a constitutively realistic finite element approach for volumetric displacement estimation. The proposed registration method integrates in a unified framework all of the stages required to estimate the movement of the cortical surface in the operating room: model-based segmentation of the pre-operative brain surface in magnetic resonance image data, range-sensing of the cortex in the OR, range-MR rigid transformation computation, and range-based non-rigid brain motion estimation. The brain segmentation technique is an adaptation of the surface evolution model. Its convergence to the brain boundary is the result of a speed term restricted to white and grey matter voxels made explicit by a classifier, and the final result is post-processed to yield a Closest Point Map of the brain surface in MR space. In turn, this Closest Point Map is used to produce the homologous pairs required to determine a highly efficient, 2D spline-based, Iterative Closest Point (ICP) non-rigid surface registration. The baseline for computing intra-operative brain displacement, as well as the initial starting point of the non-rigid ICP registration, is determined by a very good rigid range-MR transformation, produced by a simple procedure for relating the range coordinate system to that of the probe, and ultimately to that of the MR volume.

Entropy-based gaze planning

Abstract This paper describes an algorithm for recognizing known objects in an unstructured environment (e.g. landmarks) from measurements acquired with a single monochrome television camera mounted on a mobile observer. The approach is based on the concept of an entropy map , which is used to guide the mobile observer along an optimal trajectory that minimizes the ambiguity of recognition as well as the amount of data that must be gathered. Recognition itself is based on the optical flow signatures that result from the camera motion — signatures that are inherently ambiguous due to the confounding of motion, structure and imaging parameters. We show how gaze planning partially alleviates this problem by generating trajectories that maximize discriminability. A sequential Bayes approach is used to handle the remaining ambiguity by accumulating evidence for different object hypotheses over time until a clear assertion can be made. Results from an experimental recognition system using a gantry-mounted television camera are presented to show the effectiveness of the algorithm on a large class of common objects.

Recovery of Volumetric Object Descriptions From Laser Rangefinder Images

This paper describes a representation and computational model for deriving three dimensional, articulated volumetric descriptions of objects from laser rangefinder data. What differentiates this work from other approaches is that it is purely bottom-up, relying on general assumptions cast in terms of differential geometry.

Where and why local shading analysis works

An apparent contradiction in the shape-from-shading literature is examined. Although practical experience suggests that shape can be inferred from the local analysis of shading, mathematical analyses support the opposite. A criterion for exact surface recovery is derived, and it is shown that, as a result of surface geometry, elliptic, hyperbolic, and doubly curved surfaces could be recovered to reasonable accuracy provided that surface curvature and/or foreshortening were limited. In other words, by relaxing the requirement that surfaces be recovered exactly, it is found that local analysis of shading can provide useful descriptions of shape as evidenced by the results presented. >

Blind super-resolution using a learning-based approach

The super-resolution of a single image of unknown point spread-function (PSF) is addressed by extending a learning framework using blind deconvolution with an uncertainty around the resulting PSF. Results indicate success in refining the estimate of the PSF as well as to restoring the image. A novel disparity measure is also proposed to quantify the results.

Active recognition: using uncertainty to reduce ambiguity

Scene ambiguity, due to noisy measurements and uncertain object models, can be quantified and actively used by an autonomous agent to efficiently gather new data and improve its information about the environment. In this work an information-based utility measure is used to derive from a learned classification of shape models an efficient data collection strategy, specifically aimed at increasing classification confidence when recognizing uncertain shapes.

Cell Tracking: A Modeling and Minimization Approach

This paper presents a model of motion suitable for cell tracking. It includes a representation for cell dynamics enabling it to maintain a correspondence between successive images of cells undergoing morphological changes. This model is based on a minimization problem whose computational solution is similar in form to a Newton-Rhapson iteration. The model is supported by experimental results from an actual tracking problem.