Motivation: Image augmentation is a frequently used technique in computer vision and has been seeing increased interest since the popularity of deep learning. Its usefulness is becoming more and more recognized due to deep neural networks requiring larger amounts of data to train, and because in certain fields, such as biomedical imaging, large amounts of labelled data are difficult to come by or expensive to produce. In biomedical imaging, features specific to this domain need to be addressed. Results: Here we present the Augmentor software package for image augmentation. It provides a stochastic, pipeline-based approach to image augmentation with a number of features that are relevant to biomedical imaging, such as z-stack augmentation and randomized elastic distortions. The software has been designed to be highly extensible meaning an operation that might be specific to a highly specialized task can easily be added to the library, even at runtime. Although it has been designed as a general software library, it has features that are particularly relevant to biomedical imaging and the techniques required for this domain. Availability and implementation: Augmentor is a Python package made available under the terms of the MIT licence. Source code can be found on GitHub under https://github.com/mdbloice/Augmentor and installation is via the pip package manager (A Julia version of the package, developed in parallel by Christof Stocker, is also available under https://github.com/Evizero/Augmentor.jl).

We propose a scalable, efficient and accurate approach to retrieve 3D models for objects in the wild. Our contribution is twofold. We first present a 3D pose estimation approach for object categories which significantly outperforms the state-of-the-art on Pascal3D+. Second, we use the estimated pose as a prior to retrieve 3D models which accurately represent the geometry of objects in RGB images. For this purpose, we render depth images from 3D models under our predicted pose and match learned image descriptors of RGB images against those of rendered depth images using a CNN-based multi-view metric learning approach. In this way, we are the first to report quantitative results for 3D model retrieval on Pascal3D+, where our method chooses the same models as human annotators for 50% of the validation images on average. In addition, we show that our method, which was trained purely on Pascal3D+, retrieves rich and accurate 3D models from ShapeNet given RGB images of objects in the wild.

In this work, we introduce an end-to-end workflow for very high-resolution satellite-based mapping, building the basis for important 3D mapping products: (1) digital surface model, (2) digital terrain model, (3) normalized digital surface model and (4) ortho-rectified image mosaic. In particular, we describe all underlying principles for satellite-based 3D mapping and propose methods that extract these products from multi-view stereo satellite imagery. Our workflow is demonstrated for the Pléiades satellite constellation, however, the applied building blocks are more general and thus also applicable for different setups. Besides introducing the overall end-to-end workflow, we need also to tackle single building blocks: optimization of sensor models represented by rational polynomials, epipolar rectification, image matching, spatial point intersection, data fusion, digital terrain model derivation, ortho rectification and ortho mosaicing. For each of these steps, extensions to the state-of-the-art are proposed and discussed in detail. In addition, a novel approach for terrain model generation is introduced. The second aim of the study is a detailed assessment of the resulting output products. Thus, a variety of data sets showing different acquisition scenarios are gathered, allover comprising 24 Pléiades images. First, the accuracies of the 2D and 3D geo-location are analyzed. Second, surface and terrain models are evaluated, including a critical look on the underlying error metrics and discussing the differences of single stereo, tri-stereo and multi-view data sets. Overall, 3D accuracies in the range of $0.2$ to $0.3$ m in planimetry and $0.2$ to $0.4$ m in height are achieved w.r.t. ground control points. Retrieved surface models show normalized median absolute deviations around $0.9$ m in comparison to reference LiDAR data. Multi-view stereo outperforms single stereo in terms of accuracy and completeness of the resulting surface models.

We present an efficient method for geolocalization in urban environments starting from a coarse estimate of the location provided by a GPS and using a simple untextured 2.5D model of the surrounding buildings. Our key contribution is a novel efficient and robust method to optimize the pose: We train a Deep Network to predict the best direction to improve a pose estimate, given a semantic segmentation of the input image and a rendering of the buildings from this estimate. We then iteratively apply this CNN until converging to a good pose. This approach avoids the use of reference images of the surroundings, which are difficult to acquire and match, while 2.5D models are broadly available. We can therefore apply it to places unseen during training.