Moving object detection in satellite videos (SVMOD) is a challenging task due to the extremely dim and small target characteristics. Current learning-based methods extract spatio-temporal information from multi-frame dense representation with labor-intensive manual labels to tackle SVMOD, which needs high annotation costs and contains tremendous computational redundancy due to the severe imbalance between foreground and background regions. In this paper, we propose a highly efficient unsupervised framework for SVMOD. Specifically, we propose a generic unsupervised framework for SVMOD, in which pseudo labels generated by a traditional method can evolve with the training process to promote detection performance. Furthermore, we propose a highly efficient and effective sparse convolutional anchor-free detection network by sampling the dense multi-frame image form into a sparse spatio-temporal point cloud representation and skipping the redundant computation on background regions. Coping these two designs, we can achieve both high efficiency (label and computation efficiency) and effectiveness. Extensive experiments demonstrate that our method can not only process 98.8 frames per second on $1024 \times 1024$ images but also achieve state-of-the-art performance.

We study the problem of efficient object detection in 3D point clouds with the voxel-point framework. Considering a large number of redundant and dense proposals are usually generated for small-size objects during inference in voxel-based single-stage detectors, existing detectors usually introduce extra subnetworks to filter and further refine the redundancy proposals. Albeit feasible, the computational and memory cost also increase during inference. In this paper, we introduce a novel voxel-to-point 3D detector, termed as V2P-SSD, which is a novel and lightweight pipeline that jointly integrates the voxel backbone and point head together in a single-stage framework. Different from dense predictions in feature maps, voxels related to objects in our framework are sampled with a fixed number and then transformed into points. Consequently, the point head is used to dynamically generate object proposals. Our voxel-to-point detection paradigm demonstrates a significant precision improvement on small-sized objects without introducing extra memory footprints. Extensive experiments conducted on KITTI and ONCE benchmarks validate the superiority of our method.

Efficiently compressing large-scale point cloud data under limited bandwidth and computing resource conditions has become a critical issue to be addressed in mobile computing platforms. Although the octree structure can efficiently represent large-scale and complex point clouds, existing octree-based Point Cloud Geometry Compression (PCGC) approaches typically focus on exploiting either spatial or channel features individually, neglecting the interaction across spatial-channel dimensions. In addition, current approaches are also limited to small-scale point clouds due to reliance on global Transformer or local convolutional neural network (CNN). To solve these issues, we introduce GCFI-Net, a global-local cross-spatial-channel feature interaction network for predicting the occupancy probability distribution of each octree node in this paper. In the GCFI-Net, we propose a Multiscale Convolutional Fusion-based Spatial Interaction (MCFSI) module to capture global context and model spatial interactions, and a Global-Local Cross-Channel Interaction (GLCCI) module with dual pathways to integrate global and local cross-channel information. Additionally, we propose a Multiscale-enhanced Spatial and Channel Interaction (MSCI) module to aggregate features from ancestor and sibling nodes, which further enhances the octree node representation ability. Extensive experiments on large-scale sparse LiDAR and dense human body point clouds demonstrate that the proposed GCFI-Net achieves superior compression performance with fewer parameters compared to state-of-the-art PCGC methods.

The tasks of point cloud analysis are very challenging. Designing efficient convolution operation is the key to accomplish these tasks. In order to capture the structure information, neighborhood usually needs to be considered when designing convolution. At present, most of the works adopt K-Nearest Neighbor or ball query to construct neighborhood. However, these two methods only focus on the spatial distance relationship and ignore the long-distance dependence between points. In this paper, Learnable-Graph Convolutional Neural Network (LG-CNN) is proposed, which can adaptively search the backbone graph of objects. The key of LG-CNN is to design a learning-based neighborhood search method, which adaptively searches the overall backbone information of the object for each central point. Compared with updating the central point through local information aggregation, the effect of using backbone information to update the central point is better. Moreover, a graph convolution is designed to adaptively obtain the unique relationship between points and capture the diversified links between different neighbors. The challenging benchmark experiments of three tasks verify LG-CNN achieves competitive results.

Transformer plays an increasingly important role in various computer vision areas and remarkable achievements have also been made in point cloud analysis. Since they mainly focus on point-wise transformer, an adaptive channel encoding transformer is proposed in this paper. Specifically, a channel convolution called Transformer-Conv is designed to encode the channel. It can encode feature channels by capturing the potential relationship between coordinates and features. Compared with simply assigning attention weight to each channel, our method aims to encode the channel adaptively. In addition, our network adopts the neighborhood search method of low-level and high-level dual semantic receptive fields to improve the performance. Extensive experiments show that our method is superior to state-of-the-art point cloud classification and segmentation methods on three benchmark datasets.