Image-to-point cloud registration methods typically follow a coarse-to-fine pipeline, extracting patch-level correspondences and refining them into dense pixel-to-point matches. However, in scenes with repetitive patterns, images often lack sufficient 3D structural cues and alignment with point clouds, leading to incorrect matches. Moreover, prior methods usually overlook structural consistency, limiting the full exploitation of correspondences. To address these issues, we propose two novel modules: the Local Geometry Enhancement (LGE) module and the Graph Distribution Consistency (GDC) module. LGE enhances both image and point cloud features with normal vectors, injecting geometric structure into image features to reduce mismatches. GDC constructs a graph from matched points to update features and explicitly constrain similarity distributions. Extensive experiments and ablations on two benchmarks, RGB-D Scenes v2 and 7-Scenes, demonstrate that our approach achieves state-of-the-art performance in image-to-point cloud registration.

The method for image-to-point cloud registration typically determines the rigid transformation using a coarse-to-fine pipeline. However, directly and uniformly matching image patches with point cloud patches may lead to focusing on incorrect noise patches during matching while ignoring key ones. Moreover, due to the significant differences between image and point cloud modalities, it may be challenging to bridge the domain gap without specific improvements in design. To address the above issues, we innovatively propose the Uncertainty-aware Hierarchical Matching Module (UHMM) and the Adversarial Modal Alignment Module (AMAM). Within the UHMM, we model the uncertainty of critical information in image patches and facilitate multi-level fusion interactions between image and point cloud features. In the AMAM, we design an adversarial approach to reduce the domain gap between image and point cloud. Extensive experiments and ablation studies on RGB-D Scene V2 and 7-Scenes benchmarks demonstrate the superiority of our method, making it a state-of-the-art approach for image-to-point cloud registration tasks.

Image-to-point cloud registration aims to estimate the camera pose of a given image within a 3D scene point cloud. In this area, matching-based methods have achieved leading performance by first detecting the overlapping region, then matching point and pixel features learned by neural networks and finally using the PnP-RANSAC algorithm to estimate camera pose. However, achieving accurate image-to-point cloud registration remains challenging because the overlapping region detection is unreliable merely relying on point-wise classification, direct alignment of cross-modal data is difficult and indirect optimization objective leads to unstable registration results. To address these challenges, we propose a novel implicit correspondence learning method, including a Geometric Prior-guided overlapping region Detection Module (GPDM), an Implicit Correspondence Learning Module (ICLM), and a Pose Regression Module (PRM). The proposed method enjoys several merits. First, the proposed GPDM can precisely detect the overlapping region. Second, the ICLM can generate robust cross-modality correspondences. Third, the PRM can enable end-to-end optimization. Extensive experimental results on KITTI and nuScenes datasets demonstrate that the proposed model sets a new state-of-the-art performance in registration accuracy.

Detection-free methods typically follow a coarse-to-fine pipeline, extracting image and point cloud features for patch-level matching and refining dense pixel-to-point correspondences. However, differences in feature channel attention between images and point clouds may lead to degraded matching results, ultimately impairing registration accuracy. Furthermore, similar structures in the scene could lead to redundant correspondences in cross-modal matching. To address these issues, we propose Channel Adaptive Adjustment Module (CAA) and Global Optimal Selection Module (GOS). CAA enhances intra-modal features and suppresses cross-modal sensitivity, while GOS replaces local selection with global optimization. Experiments on RGB-D Scenes V2 and 7-Scenes demonstrate the superiority of our method, achieving state-of-the-art performance in image-to-point cloud registration.

Event cameras offer many advantages over traditional frame-based cameras, such as high dynamic range and low latency. Therefore, event cameras are widely applied in diverse computer vision applications, where event-based key-point detection is a fundamental task. However, achieving robust event-based keypoint detection remains challenging because the ground truth of event keypoints is difficult to obtain, descriptors extracted by CNN usually lack discrimi-native ability in the presence of intense noise, and fixed key-point detectors are limited in detecting varied keypoint patterns. To address these challenges, a novel event-based key-point detection method is proposed by learning dynamic de-tectors and contextual descriptors in a self-supervised manner (SD2Event), including a contextual feature descriptor learning (CFDL) module and a dynamic keypoint detector learning (DKDL) module. The proposed SD2Event enjoys several merits. First, the proposed CFDL module can model long-range contexts efficiently and effectively. Second, the DKDL module generates dynamic keypoint detectors, which can detect keypoints with diverse patterns across various event streams. Third, the proposed self-supervised signals can guide the model's adaptation to event data. Extensive experimental results on three challenging benchmarks show that our proposed method significantly outperforms state-of-the-art event-based keypoint detection methods.

To handle the impact of the explosive growth of data traffic generated by smart industrial applications in the Industrial Internet of Things (IIoT) scenario, edge caching is commonly used in the IIoT to reduce the content access delay and to release the backhaul load. However, due to the mobility of IIoT devices and temporal dependence of popularity, traditional content caching policies, such as least frequently used (LFU) and least recently used (LRU), cannot cache the required contents accurately within the coverage of edge server. Therefore, by exploiting the moving trajectory of the IIoT devices and the temporal dependence of the content popularity, we propose a multiagent cooperative caching policy, in which each edge server acts as an agent to cooperatively learn the optimal caching decision, then each edge server caches the corresponding contents to reduce the content access delay. Especially, we first apply the$K$-order Markov chain to predict the moving trajectory of the IIoT devices to get the IIoT device set within the coverage of each edge server. Second, we use long short-term memory (LSTM) to get the prior knowledge of the content requests by using the moving trajectory prediction results. Finally, we obtain the optimal caching decision by deep reinforcement learning to improve the Quality of Service (QoS) of the IIoT applications. Experimental results demonstrate that the proposed caching policy can efficiently improve the cache-hit ratio and content access delay.

Single-cell Hi-C (scHi-C) sequencing technologies allow us to investigate three-dimensional chromatin organization at the single-cell level. However, we still need computational tools to deal with the sparsity of the contact maps from single cells and embed single cells in a lower-dimensional Euclidean space. This embedding helps us understand relationships between the cells in different dimensions, such as cell-cycle dynamics and cell differentiation. We present an open-source computational toolbox, scHiCTools, for analyzing single-cell Hi-C data comprehensively and efficiently. The toolbox provides two methods for screening single cells, three common methods for smoothing scHi-C data, three efficient methods for calculating the pairwise similarity of cells, three methods for embedding single cells, three methods for clustering cells, and a build-in function to visualize the cells embedding in a two-dimensional or three-dimensional plot. scHiCTools, written in Python3, is compatible with different platforms, including Linux, macOS, and Windows.