We tasked two supervised keypoint detection networks to automatically identify anatomically relevant landmarks on sheep bones. The results of the models were compared to manual labeling by a domain expert, yielding satisfactory initial results that provide a promising foundation for further research.

Group Anomaly Detection (GAD) identifies unusual pattern in groups where individual members might not be anomalous. This task is of major importance across multiple disciplines, in which also sequences like trajectories can be considered as a group. As groups become more diverse in heterogeneity and size, detecting group anomalies becomes challenging, especially without supervision. Though Recurrent Neural Networks are well established deep sequence models, their performance can decrease with increasing sequence lengths. Hence, this paper introduces GADformer, a BERT-based model for attention-driven GAD on trajectories in unsupervised and semi-supervised settings. We demonstrate how group anomalies can be detected by attention-based GAD. We also introduce the Block-Attention-anomaly-Score (BAS) to enhance model transparency by scoring attention patterns. In addition to that, synthetic trajectory generation allows various ablation studies. In extensive experiments we investigate our approach versus related works in their robustness for trajectory noise and novelties on synthetic data and three real world datasets.

Bioarchaeology aims to reconstruct, e.g., the diet or provenance of animals and humans in archaeological times. This can be done by investigating so-called multi-isotope fingerprints, resulting from the analysis of several different isotope systems in parallel. However, the multi-isotope fingerprint of samples from coastal regions can be influenced by the so-called sea spray effect, resulting in "too marine" isotope signatures in terrestrial herbivorous individuals, falling in-between less or un-affected herbivores and marine mammals when clustering the isotope data ("sea spray cluster"), what cannot be explained by the diet or habitat of these individuals. The recently proposed in-between instance (IBI) definition allows the detection of additional sea spray candidates, not grouped into the sea spray cluster, thus less affected by sea spray but still of interest to domain experts. The sea spray effect locally expected in archaeological individuals can also be investigated by isotope analysis in modern plants of the same region. The fusion of isotope data measured in archaeological bones and in modern plants allows us to investigate samples (un-)affected by sea spray, what is relevant for domain experts to understand the potential local range of isotope values at an archaeological site.

The goal of the #Hyperview challenge is to use Hyperspectral Imaging (HSI) to predict the soil parameters potassium (K), phosphorus pentoxide (P2O5), magnesium (Mg) and the pH value. These are relevant parameters to determine the need of fertilization in agriculture. With this knowledge, fertilizers can be applied in a targeted way rather than in a prophylactic way which is the current procedure of choice.In this context we introduce two different approaches to solve this regression task based on 3D CNNs with Huber loss regression (SpectralNet3D) and on 1D RNNs. Both methods show distinct advantages with a peak challenge metric score of 0.808 on provided validation data.

High-quality data is necessary for modern machine learning. However, the acquisition of such data is difficult due to noisy and ambiguous annotations of humans. The aggregation of such annotations to determine the label of an image leads to a lower data quality. We propose a data-centric image classification benchmark with nine real-world datasets and multiple annotations per image to allow researchers to investigate and quantify the impact of such data quality issues. With the benchmark we can study the impact of annotation costs and (semi-)supervised methods on the data quality for image classification by applying a novel methodology to a range of different algorithms and diverse datasets. Our benchmark uses a two-phase approach via a data label improvement method in the first phase and a fixed evaluation model in the second phase. Thereby, we give a measure for the relation between the input labeling effort and the performance of (semi-)supervised algorithms to enable a deeper insight into how labels should be created for effective model training. Across thousands of experiments, we show that one annotation is not enough and that the inclusion of multiple annotations allows for a better approximation of the real underlying class distribution. We identify that hard labels can not capture the ambiguity of the data and this might lead to the common issue of overconfident models. Based on the presented datasets, benchmarked methods, and analysis, we create multiple research opportunities for the future directed at the improvement of label noise estimation approaches, data annotation schemes, realistic (semi-)supervised learning, or more reliable image collection.

Stixels have been successfully applied to a wide range of vision tasks in autonomous driving, recently including instance segmentation. However, due to their sparse occurrence in the image, until now Stixels seldomly served as input for Deep Learning algorithms, restricting their utility for such approaches. In this work we present StixelPointNet, a novel method to perform fast instance segmentation directly on Stixels. By regarding the Stixel representation as unstructured data similar to point clouds, architectures like PointNet are able to learn features from Stixels. We use a bounding box detector to propose candidate instances, for which the relevant Stixels are extracted from the input image. On these Stixels, a PointNet models learns binary segmentations, which we then unify throughout the whole image in a final selection step. StixelPointNet achieves state-of-the-art performance on Stixel-level, is considerably faster than pixel-based segmentation methods, and shows that with our approach the Stixel domain can be introduced to many new 3D Deep Learning tasks.

This contribution deals with image restoration in optical systems with coherent illumination, which is an important topic in astronomy, coherent microscopy and radar imaging. Such optical systems suffer from wavefront distortions, which are caused by imperfect imaging components and conditions. Known image restoration algorithms work well for incoherent imaging, they fail in case of coherent images. In this paper a novel wavefront correction algorithm is presented, which allows image restoration under coherent conditions. In most coherent imaging systems, especially in astronomy, the wavefront deformation is known. Using this information, the proposed algorithm allows a high quality restoration even in case of severe wavefront distortions. We present two versions of this algorithm, which are an evolution of the Gerchberg-Saxton and the Hybrid-Input-Output algorithm. The algorithm is verified on simulated and real microscopic images.