Drones have become essential tools in applications ranging from surveillance to disaster management, but their misuse poses significant security threats that require effective detection and classification techniques. While various detection methods exist, RF-based approaches offer high accuracy by using the unique spectral signatures emitted during drone communication, avoiding common issues found in visual or audio systems. However, there is significant lack of formatting and organization in public and private datasets for RF-based signals, which can hinder the use of said datasets for further research.This paper presents a comparative analysis of four RF-based drone detection datasets: DroneRF, CardRF, VTI_DroneSET_FFT, and Drone-Remote-Controller-RF-Signal-Dataset. We evaluated the experimental methodologies, including the number and types of drones used, signal features extracted, the data collection environment, the use of the RF chamber, noise mitigation techniques and classification models used. Additionally, we replicated the DroneRF analysis using CardRF data and vice versa to assess cross-dataset reproducibility, as well as investigate the robustness of the datasets concerning different RF channel conditions, variations in training and testing environments, and the overall classification accuracy. This process also revealed key challenges that hindered reproducibility and interoperability between datasets, such as inconsistent and insufficient documentation, hard-coded local file paths, and language-specific dependencies. Our study provides insight into the impact of these methodological differences on drone detection and classification performance in the real world and the importance of standardized data formats and well-documented workflows to enable consistent cross-platform RF-based drone detection research.

Deep learning with 3D data such as reconstructed point clouds and CAD models has received great research interests recently. However, the capability of using point clouds with convolutional neural network has been so far not fully explored. In this paper, we present a convolutional neural network for semantic segmentation and object recognition with 3D point clouds. At the core of our network is point-wise convolution, a new convolution operator that can be applied at each point of a point cloud. Our fully convolutional network design, while being surprisingly simple to implement, can yield competitive accuracy in both semantic segmentation and object recognition task.

Several RGB-D datasets have been publicized over the past few years for facilitating research in computer vision and robotics. However, the lack of comprehensive and fine-grained annotation in these RGB-D datasets has posed challenges to their widespread usage. In this paper, we introduce SceneNN, an RGB-D scene dataset consisting of 100 scenes. All scenes are reconstructed into triangle meshes and have per-vertex and per-pixel annotation. We further enriched the dataset with fine-grained information such as axis-aligned bounding boxes, oriented bounding boxes, and object poses. We used the dataset as a benchmark to evaluate the state-of-the-art methods on relevant research problems such as intrinsic decomposition and shape completion. Our dataset and annotation tools are available at http://www.scenenn.net.

This paper proposes a field model for repairing 3D shapes constructed from multi-view RGB data. Specifically, we represent a 3D shape in a Markov random field (MRF) in which the geometric information is encoded by random binary variables and the appearance information is retrieved from a set of RGB images captured at multiple viewpoints. The local priors in the MRF model capture the local structures of object shapes and are learnt from 3D shape templates using a convolutional deep belief network. Repairing a 3D shape is formulated as the maximum a posteriori (MAP) estimation in the corresponding MRF. Variational mean field approximation technique is adopted for the MAP estimation. The proposed method was evaluated on both artificial data and real data obtained from reconstruction of practical scenes. Experimental results have shown the robustness and efficiency of the proposed method in repairing noisy and incomplete 3D shapes.

Detecting highly articulated objects such as humans is a challenging problem. This paper proposes a novel part-based model built upon poselets, a notion of parts, and Markov Random Field (MRF) for modelling the human body structure under the variation of human poses and viewpoints. The problem of human detection is then formulated as maximum a posteriori (MAP) estimation in the MRF model. Variational mean field method, a robust statistical inference, is adopted to approximate the MAP estimation. The proposed method was evaluated and compared with existing methods on different test sets including H3D and PASCAL VOC 2007-2009. Experimental results have favourbly shown the robustness of the proposed method in comparison to the state-of-the-art.