Variational autoencoders (VAEs) play an important role in high-dimensional data generation based on their ability to fuse the stochastic data representation with the power of recent deep learning techniques. The main advantages of these types of generators lie in their ability to encode the information with the possibility to decode and generalize new samples. This capability was heavily explored for 2D image processing; however, only limited research focuses on VAEs for 3D data processing. In this article, we provide a thorough review of the latest achievements in 3D data processing using VAEs. These 3D data types are mostly point clouds, meshes, and voxel grids, which are the focus of a wide range of applications, especially in robotics. First, we shortly present the basic autoencoder with the extensions towards the VAE with further subcategories relevant to discrete point cloud processing. Then, the 3D data specific VAEs are presented according to how they operate on spatial data. Finally, a few comprehensive table summarizing the methods, codes, and datasets as well as a citation map is presented for a better understanding of the VAEs applied to 3D data. The structure of the analyzed papers follows a taxonomy, which differentiates the algorithms according to their primary data types and application domains.

A novel method is proposed for the absolute pose estimation of a central 2D camera with respect to 3D depth data without the use of any dedicated calibration pattern or explicit point correspondences. The proposed method has no specific assumption about the data source: plain depth information is expected from the 3D sensing device and a central camera is used to capture the 2D images. Both the perspective and omnidirectional central cameras are handled within a single generic camera model. Pose estimation is formulated as a 2D-3D nonlinear shape registration task which is solved without point correspondences or complex similarity metrics. It relies on a set of corresponding planar regions, and the pose parameters are obtained by solving an overdetermined system of nonlinear equations. The efficiency and robustness of the proposed method were confirmed on both large scale synthetic data and on real data acquired from various types of sensors.

In this paper, an overview and a demonstration for the vertical integration of manufacturing enterprise layers are described by implementing the manufacturing execution system (MES). In the first part of the paper, the details regarding the MES implementation are described, while in the second part, the use case-specific insights are highlighted. The presented use case contains each important step of a production line involving also a collaborative Baxter-type robot and the state-of-the-art tools for MES implementation. The cobot-involved use case is relevant and generic enough in the context of Industry 4.0 offering a good overview of a typical vertical integration use case which can be generalised and applied to manufacturing scenarios. The paper ends with the lessons learned from the vertical integration process as well as future direction which can be followed in such a context.

Vine disease detection is considered one of the most crucial components in precision viticulture. It serves as an input for several further modules, including mapping, automatic treatment, and spraying devices. In the last few years, several approaches have been proposed for detecting vine disease based on indoor laboratory conditions or large-scale satellite images integrated with machine learning tools. However, these methods have several limitations, including laboratory-specific conditions or limited visibility into plant-related diseases. To overcome these limitations, this work proposes a low-altitude drone flight approach through which a comprehensive dataset about various vine diseases from a large-scale European dataset is generated. The dataset contains typical diseases such as downy mildew or black rot affecting the large variety of grapes including Muscat of Hamburg, Alphonse Lavallée, Grasă de Cotnari, Rkatsiteli, Napoca, Pinot blanc, Pinot gris, Chambourcin, Fetească regală, Sauvignon blanc, Muscat Ottonel, Merlot, and Seyve-Villard 18402. The dataset contains 10,000 images and more than 100,000 annotated leaves, verified by viticulture specialists. Grape bunches are also annotated for yield estimation. Further, tests were made against state-of-the-art detection methods on this dataset, focusing also on viable solutions on embedded devices, including Android-based phones or Nvidia Jetson boards with GPU. The datasets, as well as the customized embedded models, are available on the project webpage.2

Leukemia is a deadly disease, and the patient’s recovery rate is very dependent on early diagnosis. However, its diagnosis under the microscope is tedious and time-consuming. The advancement of deep convolutional neural networks (CNNs) in image classification has enabled new techniques in automated disease detection systems. These systems serve as valuable support and secondary opinion resources for laboratory technicians and hematologists when diagnosing leukemia through microscopic examination. In this study, we deployed a pre-trained CNN model (MobileNet) that has a small size and low complexity, making it suitable for mobile applications and embedded systems. We used the L1 regularization method and a novel dataset balancing approach, which incorporates HSV color transformation, saturation elimination, Gaussian noise addition, and several established augmentation techniques, to prevent model overfitting. The proposed model attained an accuracy of 95.33% and an F1 score of 0.95 when evaluated on the held-out test set extracted from the C_NMC_2019 public dataset. We also evaluated the proposed model by adding zero-mean Gaussian noise to the test images. The experimental results indicate that the proposed model is both efficient and robust, even when subjected to additional Gaussian noise. The comparison of the proposed MobileNet_M model’s results with those of ALNet and various other existing models on the same dataset underscores its superior efficacy. The code is available for reproducing the experimental results at https://tamaslevente.github.io/ALLM/.