We present a conceptually simple framework for object instance segmentation, called Contour Proposal Network (CPN), which detects possibly overlapping objects in an image while simultaneously fitting closed object contours using a fixed-size representation based on Fourier Descriptors. The CPN can incorporate state-of-the-art object detection architectures as backbone networks into a single-stage instance segmentation model that can be trained end-to-end. We construct CPN models with different backbone networks and apply them to instance segmentation of cells in datasets from different modalities. In our experiments, CPNs outperform U-Net, Mask R-CNN and StarDist in instance segmentation accuracy. We present variants with execution times suitable for real-time applications. The trained models generalize well across different domains of cell types. Since the main assumption of the framework is closed object contours, it is applicable to a wide range of detection problems also beyond the biomedical domain. An implementation of the model architecture in PyTorch is freely available.

The use of Deep Learning methods have been identified as a key opportunity for enabling processing of extreme-scale scientific datasets. Feeding data into compute nodes equipped with several high-end GPUs at sufficiently high rate is a known challenge. Facilitating processing of these datasets thus requires the ability to store petabytes of data as well as to access the data with very high bandwidth. In this work, we look at two Deep Learning use cases for cytoarchitectonic brain mapping. These applications are very challenging for the underlying IO system. We present an in depth analysis of their IO requirements and performance. Both applications are limited by the IO performance, as the training processes often have to wait several seconds for new training data. Both applications read random patches from a collection of large HDF5 datasets or TIFF files, which result in many small non-consecutive accesses to the parallel file systems. By using a chunked data format or storing temporally copies of the required patches, the IO performance can be improved significantly. These leads to a decrease of the total runtime of up to 80%.

We present a scheme for measuring completeness of local feature extraction in terms of image coding. Completeness is here considered as good coverage of relevant image information by the features. As each feature requires a certain number of bits which are representative for a certain subregion of the image, we interpret the coverage as a sparse coding scheme. The measure is therefore based on a comparison of two densities over the image domain: An entropy density pH(x) based on local image statistics, and a feature coding density pc(x) which is directly computed from each particular set of local features. Motivated by the coding scheme in JPEG, the entropy distribution is derived from the power spectrum of local patches around each pixel position in a statistically sound manner. As the total number of bits for coding the image and for representing it with local features may be different, we measure incompleteness by the Hellinger distance between pH(x) and pc(x). We will derive a procedure for measuring incompleteness of possibly mixed sets of local features and show results on standard datasets using some of the most popular region and keypoint detectors, including Lowe, MSER and the recently published SFOP detectors. Furthermore, we will draw some interesting conclusions about the complementarity of detectors.

This paper presents a novel method for detecting scale invariant keypoints. It fills a gap in the set of available methods, as it proposes a scale-selection mechanism for junction-type features. The method is a scale-space extension of the detector proposed by Förstner (1994) and uses the general spiral feature model of Bigün (1990) to unify different types of features within the same framework. By locally optimising the consistency of image regions with respect to the spiral model, we are able to detect and classify image structures with complementary properties over scale-space, especially star and circular shapes as interpretable and identifiable subclasses. Our motivation comes from calibrating images of structured scenes with poor texture, where blob detectors alone cannot find sufficiently many keypoints, while existing corner detectors fail due to the lack of scale invariance. The procedure can be controlled by semantically clear parameters. One obtains a set of keypoints with position, scale, type and consistency measure. We characterise the detector and show results on common benchmarks. It competes in repeatability with the Lowe detector, but finds more stable keypoints in poorly textured areas, and shows comparable or higher accuracy than other recent detectors. This makes it useful for both object recognition and camera calibration.