Christoph Dann

Bayesian Time-of-Flight for Realtime Shape, Illumination and Albedo

We propose a computational model for shape, illumination and albedo inference in a pulsed time-of-flight (TOF) camera. In contrast to TOF cameras based on phase modulation, our camera enables general exposure profiles. This results in added flexibility and requires novel computational approaches. To address this challenge we propose a generative probabilistic model that accurately relates latent imaging conditions to observed camera responses. While principled, realtime inference in the model turns out to be infeasible, and we propose to employ efficient non-parametric regression trees to approximate the model outputs. As a result we are able to provide, for each pixel, at video frame rate, estimates and uncertainty for depth, effective albedo, and ambient light intensity . These results we present are state-of-the-art in depth imaging. The flexibility of our approach allows us to easily enrich our generative model. We demonstrate this by extending the original single-path model to a two-path model, capable of describing some multipath effects. The new model is seamlessly integrated in the system at no additional computational cost. Our work also addresses the important question of optimal exposure design in pulsed TOF systems. Finally, for benchmark purposes and to obtain realistic empirical priors of multipath and insights into this phenomena, we propose a physically accurate simulation of multipath phenomena.

Pottics – The Potts Topic Model for Semantic Image Segmentation

We present a novel conditional random field (CRF) for semantic segmentation that extends the common Potts model of spatial coherency with latent topics, which capture higher-order spatial relations of segment labels. Specifically, we show how recent approaches for producing sets of figure-ground segmentations can be leveraged to construct a suitable graph representation for this task. The CRF model incorporates such proposal segmentations as topics, modelling the joint occurrence or absence of object classes. The resulting model is trained using a structured large margin approach with latent variables. Experimental results on the challenging VOC’10 dataset demonstrate significant performance improvements over simpler models with less spatial structure.

Automated matching of pipeline corrosion features from in-line inspection data

The integrity assessment of corroded pipelines is often based on in-line inspection (ILI) results. Before determining the corrosion growth for the integrity assessment, the detected corrosion features from two or more ILIs need to be matched with respect to their location in the pipeline. The objective of this paper is to introduce a framework for automated feature matching. The input for the framework is the locations of all detected corrosion features and girth welds from each ILI. Using a multi-step approach, the size of several ILIs with a possibly large number of features is reduced to a set of independent smaller problems to match efficiently the corrosion features. The results include the matched features for the subsequent corrosion growth analysis and the identification of outliers that cannot be matched. The applied probabilistic matching assigns to each feature pair a probability of being a match to reflect the inherent uncertainty in the matching process. The proposed framework replaces manual matching, which can be time intensive and prone to errors, particularly for internal corrosion with high feature densities. It reliably matches features in pipelines and supports the integrity and risk assessment of pipeline systems.

Sample Efficient Policy Search for Optimal Stopping Domains

Optimal stopping problems consider the question of deciding when to stop an observation-generating process in order to maximize a return. We examine the problem of simultaneously learning and planning in such domains, when data is collected directly from the environment. We propose GFSE, a simple and flexible model-free policy search method that reuses data for sample efficiency by leveraging problem structure. We bound the sample complexity of our approach to guarantee uniform convergence of policy value estimates, tightening existing PAC bounds to achieve logarithmic dependence on horizon length for our setting. We also examine the benefit of our method against prevalent model-based and model-free approaches on 3 domains taken from diverse fields.