Robust Real-Time Tracking of Axis-Symmetric Magnets via Neural Networks

Traditional tracking of magnetic markers leads to high computational costs due to the requirement for iterative optimization procedures. Furthermore, such approaches rely on the magnetic dipole model for the optimization function, leading to inaccurate results anytime a non-spherical magnet gets close to a sensor in the array. We propose to overcome these limitations by using neural networks to infer the marker's position and orientation. Our method can obtain the magnet's five degrees of freedom (5 DoF) in a single inference step without relying on an initial estimation. As a counterpart, the supervised training phase is data intensive. We solve this by generating synthetic, yet realistic, data via Finite Element Methods simulations. A fast and accurate inference largely compensates for the offline training preparation. We evaluate our system using different cylindrical magnets, tracked with a square array of 16 sensors. We use a portable, neural networks-oriented single-board computer for the sensors' reading and the position inference, making our setup very compact. We compared our tracking outputs with vision-based ground truth data. Our prototype implementation tracks $5$ DoF with an averaged positional error of 4mm and orientation error of 8deg within a 0.2x0.2x0.15m working volume.

MLRSNet: A Multi-label High Spatial Resolution Remote Sensing Dataset for Semantic Scene Understanding

To better understand scene images in the field of remote sensing, multi-label annotation of scene images is necessary. Moreover, to enhance the performance of deep learning models for dealing with semantic scene understanding tasks, it is vital to train them on large-scale annotated data. However, most existing datasets are annotated by a single label, which cannot describe the complex remote sensing images well because scene images might have multiple land cover classes. Few multi-label high spatial resolution remote sensing datasets have been developed to train deep learning models for multi-label based tasks, such as scene classification and image retrieval. To address this issue, in this paper, we construct a multi-label high spatial resolution remote sensing dataset named MLRSNet for semantic scene understanding with deep learning from the overhead perspective. It is composed of high-resolution optical satellite or aerial images. MLRSNet contains a total of 109,161 samples within 46 scene categories, and each image has at least one of 60 predefined labels. We have designed visual recognition tasks, including multi-label based image classification and image retrieval, in which a wide variety of deep learning approaches are evaluated with MLRSNet. The experimental results demonstrate that MLRSNet is a significant benchmark for future research, and it complements the current widely used datasets such as ImageNet, which fills gaps in multi-label image research. Furthermore, we will continue to expand the MLRSNet. MLRSNet and all related materials have been made publicly available at https://data.mendeley.com/datasets/7j9bv9vwsx/2 and https://github.com/cugbrs/MLRSNet.git.