Search

Many timely computer vision problems, such as crowd event detection, individual or crowd activity recognition, person detection and re-identification, tracking, pose estimation, segmentation, require pixel-level annotations. This involves significant manual effort, and is likely to face challenges related to the privacy of individuals, due to the intrinsic nature of these problems, requiring in-depth identifying information. To cover the gap in the field and address these issues, we introduce and make publicly available a photorealistic, synthetically generated dataset, with detailed dense annotations. We also publish the tool we developed to generate it, that will allow users to not only use our dataset, but expand upon it by building their own densely annotated videos for many other computer vision problems. We demonstrate the usefulness of the dataset with experiments on unsupervised crowd anomaly detection in various scenarios, environments, lighting, weather conditions. Our dataset and the annotations provided with it allow its use in numerous other computer vision problems, such as pose estimation, person detection, segmentation, re-identification and tracking, individual and crowd activity recognition, and abnormal event detection. We present the dataset as is, along with the source code and tool to generate it, so any modification can be made and new data can be created. To our knowledge, there is currently no other photorealistic, densely annotated, realistic, synthetically generated dataset for abnormal crowd event detection, nor one that allows for flexibility of use by allowing the creation of new data with annotations for many other computer vision problems. Dataset and source code available: https://github.com/RicoMontulet/GTA5Event.

Biomaterial scaffolds are a key part of cardiac tissue engineering therapies. Our group has recently synthesised a novel polycaprolactone based polyurethane-urea copolymer that shows improved mechanical properties compared to its previously published counterparts. The aim of this study was to explore whether indirect 3D printing could provide a means to fabricate this novel, biodegradable polymer into a scaffold suitable for cardiac tissue engineering. Indirect 3D printing was carried out through printing water dissolvable poly(vinyl alcohol) porogens in three different sizes based on a wood-stack model, into which a polyurethane-urea solution was pressure injected. The porogens were removed, leading to soft polyurethane-urea scaffolds with regular tubular pores. The scaffolds were characterised for their compressive and tensile mechanical behaviour; and their degradation was monitored for 12 months under simulated physiological conditions. Their compatibility with cardiac myocytes and performance in novel cardiac engineering-related techniques, such as aggregate seeding and bi-directional perfusion, was also assessed. The scaffolds were found to have mechanical properties similar to cardiac tissue, and good biocompatibility with cardiac myocytes. Furthermore, the incorporated cells preserved their phenotype with no signs of de-differentiation. The constructs worked well in perfusion experiments, showing enhanced seeding efficiency. This article is protected by copyright. All rights reserved.