Many computer vision techniques infer properties of our physical world from images.
Although images are formed through the physics of light and mechanics, computer vision …

Data is the driving force of machine learning, with the amount and quality of training data
often being more important for the performance of a system than architecture and training …

BlenderProc2 is a procedural pipeline that can render realistic images for the training of
neural networks. Our pipeline can be employed in various use cases, including …

The visual organ is important for animals to obtain information and understand the outside
world; however, robots cannot do so without a visual system. At present, the vision …

BlenderProc is an open-source and modular pipeline for rendering photorealistic images of
procedurally generated 3D scenes which can be used for training data-hungry deep …

We introduce Adaptive Procedural Task Generation (APT-Gen), an approach to
progressively generate a sequence of tasks as curricula to facilitate reinforcement learning …

We present an approach for estimating a mobile robot's pose wrt the allocentric coordinates
of a network of static cameras using multi-view RGB images. The images are processed …

Dynamic scene graphs generated from video clips could help enhance the semantic visual
understanding in a wide range of challenging tasks such as environmental perception …

When training data is scarce, the incorporation of additional prior knowledge can assist the
learning process. While it is common to initialize neural networks with weights that have …

Recent problems in robotics can sometimes only be tackled using machine learning
technologies, particularly those that utilize deep learning (DL) with transfer learning …