Causal discovery—the systematic identification of cause-and-effect relationships among variables—forms the cornerstone of causal inference. Its application enables reliable predictions and targeted interventions across complex systems, from medical treatments to engineering processes. Traditional statistical causal discovery methods face significant limitations with high-dimensional data structures, while existing knowledge-based approaches rely on single large-scale models that raise fundamental concerns about computational efficiency and result reliability. The Agentic Stream of Thought (ASoT) addresses these limitations through a novel architecture that orchestrates multiple smaller open-source language models. The framework integrates hierarchical query decomposition with Model Compiler refinement, while dual-stream thought processing enables balanced analysis through parallel evaluation of competing hypotheses. Dedicated Direction and Transitive Processors enhance reasoning by resolving bidirectional relationships and refining transitive pathways. A two-tiered quality gate system and complementary consensus mechanisms—Delphi protocol and Ensemble Synthesis Method—iteratively refine outputs while mitigating hallucination risks. Empirical evaluations across causal discovery benchmarks and question-answering tasks demonstrate that this approach matches or exceeds state-of-the-art models while enabling local deployment, establishing that sophisticated orchestration of smaller models provides a more sustainable path than increasing model scale alone.

Estimating 6D object poses is a major challenge in 3D computer vision. Building on successful instance-level approaches, research is shifting towards category-level pose estimation for practical applications. Current category-level datasets, however, fall short in annotation quality and pose variety. Addressing this, we introduce HouseCat6D, a new category-level 6D pose dataset. It features 1) multi-modality with Polarimetric RGB and Depth (RGBD+P), 2) encompasses 194 diverse objects across 10 household cat-egories, including two photometrically challenging ones, and 3) provides high-quality pose annotations with an error range of only 1.35 mm to 1.74 mm. The dataset also includes 4) 41 large-scale scenes with comprehensive view-point and occlusion coverage,5) a checkerboard-free en-vironment, and 6) dense 6D parallel-jaw robotic grasp annotations. Additionally, we present benchmark results for leading category-level pose estimation networks.

Object pose estimation is crucial for robotic applications and augmented reality. Beyond instance level 6D object pose estimation methods, estimating category-level pose and shape has become a promising trend. As such, a new research field needs to be supported by well-designed datasets. To provide a benchmark with high-quality ground truth annotations to the community, we introduce a multimodal dataset for category-level object pose estimation with photometrically challenging objects termed PhoCaL. PhoCaL comprises 60 high quality 3D models of household objects over 8 categories including highly reflective, transparent and symmetric objects. We developed a novel robot-supported multi-modal (RGB, depth, polarisation) data acquisition and annotation process. It ensures sub-millimeter accuracy of the pose for opaque textured, shiny and transparent objects, no motion blur and perfect camera synchronisation. To set a benchmark for our dataset, state-of-the-art RGB-D and monocular RGB methods are evaluated on the challenging scenes of PhoCaL.

The predicted drop in prices for automotive sensors and their increasing demand are putting pressure on sensor suppliers. One possible solution is to reduce production costs by expanding automation. Nevertheless, visual quality control in particular is a process step that is often performed by human inspectors, even in the age of Industry 4.0. Software solutions can currently not be used for all types of sensor assembly quality control. This is mainly due to the difficulty of detecting both structural and logical errors and evaluating their severity. We present a machine learning-based software framework that is able to mimic the methodical behavior of a human in error detection and assessment. The framework is based on three types of models, an object recognition model, an anomaly detection model, and a segmentation model. All models are based on convolutional neural networks. An initial proof of concept (PoC) has been performed to prove the usefulness of the models and shows promising results. The initial anomaly detection model is able to reduce the number of objects to be manually tested by 16%. The object detection and segmentation are still in progress and could not be evaluated yet. In addition, a dataset preparation method is presented to use data from industrial practice and relabel it with information from an inspector survey.

The ability to successfully grasp objects is crucial in robotics, as it enables several interactive downstream applications. To this end, most approaches either compute the full 6D pose for the object of interest or learn to predict a set of grasping points. While the former approaches do not scale well to multiple object instances or classes yet, the latter require large annotated datasets and are hampered by their poor generalization capabilities to new geometries. To overcome these shortcomings, we propose to teach a robot how to grasp an object with a simple and short human demonstration. Hence, our approach neither requires many annotated images nor is it restricted to a specific geometry. We first present a small sequence of RGB-D images displaying a human-object interaction. This sequence is then leveraged to build associated hand and object meshes that represent the depicted interaction. Subsequently, we complete missing parts of the reconstructed object shape and estimate the relative transformation between the reconstruction and the visible object in the scene. Finally, we transfer the a-priori knowledge from the relative pose between object and human hand with the estimate of the current object pose in the scene into necessary grasping instructions for the robot. Exhaustive evaluations with Toyota’s Human Support Robot (HSR) in real and synthetic environments demonstrate the applicability of our proposed methodology and its advantage in comparison to previous approaches.

Driven by deep learning, object recognition has recently made a tremendous leap forward. Nonetheless, its accuracy often still suffers from several sources of variation that can be found in real-world images. Some of the most challenging variations are induced by changing lighting conditions. This paper presents a novel approach for tackling brightness variation in the domain of 2D object detection and 6D object pose estimation. Existing works aiming at improving robustness towards different lighting conditions are often grounded on classical computer vision contrast normalisation techniques or the acquisition of large amounts of annotated data in order to achieve invariance during training. While the former cannot generalise well to a wide range of illumination conditions, the latter is neither practical nor scalable. Hence, We propose the usage of Generative Adversarial Networks in order to learn how to normalise the illumination of an input image. Thereby, the generator is explicitly designed to normalise illumination in images so to enhance the object recognition performance. Extensive evaluations demonstrate that leveraging the generated data can significantly enhance the detection performance, outperforming all other state-of-the-art methods. We further constitute a natural extension focusing on white balance variations and introduce a new dataset for evaluation.