Semantic Segmentation is a crucial component in the perception systems of many applications, such as robotics and autonomous driving that rely on accurate environmental perception and understanding. In literature, several approaches are introduced to attempt LiDAR semantic segmentation task, such as projection-based (range-view or birds-eye-view), and voxel-based approaches. However, they either abandon the valuable 3D topology and geometric relations and suffer from information loss introduced in the projection process or are inefficient. Therefore, there is a need for accurate models capable of processing the 3D driving-scene point cloud in 3D space. In this paper, we propose S3Net, a novel convolutional neural network for LiDAR point cloud semantic segmentation. It adopts an encoder-decoder backbone that consists of Sparse Intra-channel Attention Module (SIntraAM), and Sparse Interchannel Attention Module (SInterAM) to emphasize the fine details of both within each feature map and among nearby feature maps. To extract the global contexts in deeper layers, we introduce Sparse Residual Tower based upon sparse convolution that suits varying sparsity of LiDAR point cloud. In addition, geo-aware anisotropic loss is leveraged to emphasize the semantic boundaries and penalize the noise within each predicted regions, leading to a robust prediction. Our experimental results show that the proposed method leads to a large improvement (12%) compared to its baseline counterpart (MinkNet42 [1]) on SemanticKITTI [2] test set and achieves state-of-the-art mIoU accuracy of semantic segmentation approaches.

No abstract

Abstract. With the advent of digital technology, digital image has gradually taken the place of the original analog photograph, and the forgery of digital image has become more and more easy and indiscoverable. Image splicing is a commonly used technique in image tampering. In this paper, we simply introduce the definition of image splicing and some methods of image splicing detection, mainly including the detection based on steganalysis model, the detection based on Hilbert-Huang transform (HHT) and moments of characteristic functions (CF) with wavelet decomposition. We focus on discussing our proposed approach based on image quality metrics (IQMs) and moment features. Especially we analyze the model creation and the extraction of features in digital image. In addition, we compare these approaches and analyze the future works of digital image forensics.

FeNC catalysts demonstrate remarkable activity and stability for the oxygen reduction reaction (ORR) in polymer electrolyte membrane fuel cells and Zn–air batteries (ZABs). The local coordination of Fe single atoms in FeNC catalysts strongly impacts ORR activity. Herein, FeNC catalysts containing Fe single atoms sites with FeN3, FeN4, and FeN5 coordinations are synthesized by carbonization of Fe‐rich polypyrrole precursors. The FeN5 sites possess a higher Fe oxidation state (+2.62) than the FeN3 (+2.23) and FeN4 (+2.47) sites, and higher ORR activity. Density functional theory calculations verify that the FeN5 coordination optimizes the adsorption and desorption of ORR intermediates, dramatically lowering the energy barrier for OH− desorption in the rate‐limiting ORR step. A primary ZAB constructed using the FeNC catalyst with FeN5 sites demonstrates state‐of‐the‐art performance (an open circuit potential of 1.629 V, power density of 159 mW cm−2). Results confirm an intimate structure‐activity relationship between Fe coordination, Fe oxidation state, and ORR activity in FeNC catalysts.