Correspondence matching is a fundamental and crucial task in robot vision. In recent years, deep learning-based keypoint matching techniques have shown outstanding performance in downstream tasks. Conventional learning-based correspondence matching methods rely on large datasets and a specific training procedure. Correspondence techniques based on pre-trained features have been preliminarily explored by researchers. Unfortunately, traditional convolutional neural networks only possess translation invariance but lack rotational invariance, hence, their performance suffers significantly under heavy rotations. Therefore, we propose a correspondence matching method based on pre-trained group-equivariant neural networks and compare the performance of various rotation-equivariant to rotation-invariant transformers. We conducted experiments on the Rotated-Hpatches and Rotated-MegaDepth datasets, and the results indicate that our proposed method is concise and effective, achieving state-of-the-art performance without the need for retraining in downstream tasks.

We propose a Visual Place Recognition (VPR) framework by sharing lightweight keypoint extraction modules for local features. Current research on the joint learning of local keypoint matching and VPR is relatively scarce, and the application deployment of real-time spatial computing on edge devices has a high learning cost. There is also a significant spatial structural difference between existing VPR methods and the scenarios in practical applications. To address these issues, we design a joint learning framework for local keypoint extraction and VPR, which shares local features and fuses irregularly distributed key features in space through self-attention and cross-attention mechanisms. Our framework achieves excellent results on several VPR datasets. In particular, we introduce a new VPR dataset, called TJPark, which has a significant spatial information difference from common street view data. Our method demonstrates that local features with strong generalization capabilities effectively help enhance the generalization of VPR. Our open source code and dataset are available at: https://github.com/ShuaiAlger/LGPR.

The rapidly evolving Internet of Things (IoT) continues to serve as a critical bridge between the physical world and digital space, driving increasing demands for optimized communication capabilities and time-sensitive data acquisition. However, traditional cloud computing architectures are becoming increasingly insufficient to meet these stringent demands. Mobile Edge Computing (MEC) has thus emerged as a promising paradigm. By delegating data processing tasks from centralized cloud servers to edge nodes (ENs) located near end-users, MEC significantly enhances service efficiency while simultaneously introducing heightened privacy and security risks. To mitigate both external threats during task interactions and internal adversaries within the network, this paper proposes a Lightweight Verification and Protection (LVP) strategy for ENs. LVP primarily leverages secure computation techniques to ensure the anonymity and confidentiality of private information during transmission and verification. It further incorporates a reputation-based evaluation framework to mitigate the impact of insider threats. Specifically, the integrity of user-uploaded task data is first verified, followed by anonymous matching between users and ENs through paired index. A reputation-based anonymous authentication algorithm is then designed to prevent exposure or linkage of reputation information during usage. Finally, reputation values are dynamically updated by jointly considering temporal relevance and historical behavior. Theoretical analysis confirms the robustness and correctness of all LVP sub-algorithms, while experimental evaluations demonstrate the scheme’s effectiveness and feasibility under low computational and communication overhead.

Monocular Bird’s Eye View (BEV) semantic segmentation is critical for autonomous driving for its inherent advantages in spatial representation and downstream tasks. However, it is challenging to simultaneously learn view transformation and pixel-wise classification. Previous works suffer from non-flat region distortion, distant depth ambiguity, and visual occlusion. To address these aforementioned concerns, we propose dual-branch view transformation (DVT), a novel framework for monocular BEV semantic segmentation. Our method consists of: (i) A dual-branch view transformation to decouple features into flat region and non-flat region and process them independently. (ii) A depth-aware weighting method to make the model pay more attention to the distant depth. (iii) An auxiliary task to introduce more inductive biases to alleviate the inaccuracy caused by visual occlusion. Furthermore, we design a class-aware weighting method to address the class and size imbalance of datasets. Experimental results on nuScenes and KITTI-360 datasets demonstrate that DVT outperforms previous state-of-the-art (SOTA). Our codes are available at https://github.com/MrPicklesGG/DVT.

In the context of 3D scene perception tasks, the significance of 3D occupancy prediction has been progressively growing, aiming to forecast the occupancy state of voxels in a discrete 3D space. However, existing methods typically exhibit several limitations, such as restricted adaptability to non-pinhole cameras due to fixed camera parameters, heavy reliance on 3D annotations because of the inability to project 3D output back to the camera plane, and inferior real-time inference performance resulting from the conversion process from 2D to 3D features. To address these constrains, we introduce GenerOcc, a self-supervised framework of real-time 3D occupancy prediction for monocular generic cameras. We have collected the fisheye Dominant dataset to confirm the compatibility of our ray-based camera model with non-pinhole cameras. By transforming the occupancy prediction task into a depth estimation task in a self-supervised manner, we eliminate dependency on 3D annotations. Furthermore, we propose a parametric voxel probability distribution module that leverages 2D features to quickly predict 3D occupancy without 3D representations of the scene. Additionally, our GenerOcc has been extensively evaluated on public pinhole Occ3D-nuScenes dataset and our proprietary fisheye Dominant dataset, both yielding impressive performance.