High-level synthesis (HLS) is an increasingly popular approach in electronic design automation (EDA) that raises the abstraction level for designing digital circuits. With the increasing complexity of embedded systems, these tools are particularly relevant in embedded systems design. In this paper, we present our evaluation of a broad selection of recent HLS tools in terms of capabilities, usability and quality of results. Even though HLS tools are still lacking some maturity, they are constantly improving and the industry is now starting to adopt them into their design flows.

In this paper, we investigate whether fusing depth information on top of normal RGB data for camera-based object detection can help to increase the performance of current state-of-the-art single-shot detection networks. Indeed, depth sensing is easily acquired using depth cameras such as a Kinect or stereo setups. We investigate the optimal manner to perform this sensor fusion with a special focus on lightweight single-pass convolutional neural network (CNN) architectures, enabling real-time processing on limited hardware. For this, we implement a network architecture allowing us to parameterize at which network layer both information sources are fused together. We performed exhaustive experiments to determine the optimal fusion point in the network, from which we can conclude that fusing towards the mid to late layers provides the best results. Our best fusion models significantly outperform the baseline RGB network in both accuracy and localization of the detections.

No abstract

In this paper we investigate the benefit of using depth information on top of normal RGB for camera-based pedestrian detection. Indeed, depth sensing is easily acquired using depth cameras such as a Kinect or stereo setups. We investigate the best way to perform this sensor fusion with a special focus on lightweight single-pass CNN architectures, enabling real-time processing on limited hardware. We implement different network architectures, each fusing depth at different layers of our network. Our experiments show that midway fusion performs the best, outperforming a regular RGB detector substantially in accuracy. Moreover, we prove that our fusion network is better at detecting individuals in a crowd, by demonstrating that it has both a better localization of pedestrians and is better at handling occluded persons. The resulting network is computationally efficient and achieves real-time performance on both desktop and embedded GPUs.

The assessment of gaze behaviour is essential for understanding the psychology of communication. Mobile eye-tracking glasses are useful to measure gaze behaviour during dynamic interactions. Eye-tracking data can be analysed by using manually annotated areas-of-interest. Computer vision algorithms may alternatively be used to reduce the amount of manual effort, but also the subjectivity and complexity of these analyses. Using additional re-identification (Re-ID) algorithms, different participants in the interaction can be distinguished. The aim of this study was to compare the results of manual annotation of mobile eye-tracking data with the results of a computer vision algorithm. We selected the first minute of seven randomly selected eye-tracking videos of consultations between physicians and patients in a Dutch Internal Medicine out-patient clinic. Three human annotators and a computer vision algorithm annotated mobile eye-tracking data, after which interrater reliability was assessed between the areas-of-interest annotated by the annotators and the computer vision algorithm. Additionally, we explored interrater reliability when using lengthy videos and different area-of-interest shapes. In total, we analysed more than 65 min of eye-tracking videos manually and with the algorithm. Overall, the absolute normalized difference between the manual and the algorithm annotations of face-gaze was less than 2%. Our results show high interrater agreements between human annotators and the algorithm with Cohen’s kappa ranging from 0.85 to 0.98. We conclude that computer vision algorithms produce comparable results to those of human annotators. Analyses by the algorithm are not subject to annotator fatigue or subjectivity and can therefore advance eye-tracking analyses.