J-SuperPoint:基于差分信息的关键点检测方法

谢仕烁; 肖泽辉; 陶杰

doi:10.12052/gdutxb.250027

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

J-SuperPoint:基于差分信息的关键点检测方法

信息与通信工程 | 更新时间：2026-05-27

- J-SuperPoint:基于差分信息的关键点检测方法
- J-SuperPoint: Keypoint Detection Method Based on Differential Information
- 广东工业大学学报 2025年42卷第3期页码：92-100
- 作者机构：
  
  广东工业大学自动化学院,广东,广州,510006
- 作者简介：
  
  谢仕烁(2000–) ，男，硕士研究生，主要研究方向为图像处理、深度学习等，E-mail：shishuoxie@foxmail.com
  陶杰(1990–)，男，副教授，博士，主要研究方向为无人自主系统、图像识别等，E-mail：taojiedyx@163.com
- 基金信息：
  
  国家自然科学基金资助项目(U23A20390，62276069);广东省基础与应用基础研究基金自然科学基金资助项目(2024B1515020094)
- DOI：10.12052/gdutxb.250027
  中图分类号： TP391.4
- 纸质出版：2025
- 稿件说明：
移动端阅览
谢仕烁, 肖泽辉, 陶杰. J-SuperPoint:基于差分信息的关键点检测方法[J]. 广东工业大学学报, 2025, 42(3): 92–100.

Xie Shishuo, Xiao Zehui, Tao Jie. J-SuperPoint: Keypoint Detection Method Based on Differential Information[J]. 2025, 42(3): 92-100.
谢仕烁, 肖泽辉, 陶杰. J-SuperPoint:基于差分信息的关键点检测方法[J]. 广东工业大学学报, 2025, 42(3): 92–100. DOI： 10.12052/gdutxb.250027

Xie Shishuo, Xiao Zehui, Tao Jie. J-SuperPoint: Keypoint Detection Method Based on Differential Information[J]. 2025, 42(3): 92-100. DOI： 10.12052/gdutxb.250027.

摘要

针对主流关键点检测算法耗时长的问题，本文提出了一种关键点检测和描述方法。首先设计了旋转差分卷积模块对图像中的差分信息进行对称性采样，随后对其进一步拓展以实现判别器模块，对差分信息和卷积信息进行采样整合，从原始图像中提取有效特征；最后，本文设计了共享编码器和轻量解码器，对解码器之间的信息进行联动，同时改进了损失函数，搭建出轻量高效的编解码网络模型。实验表明，本方法在保持较好实时性的同时，仍然具备良好的计算精度，为图像匹配、视觉定位等方法提供了兼顾实时性与精度的解决方案。

Abstract

Addressing the time-consuming issue of mainstream keypoint detection algorithms

a novel keypoint detection and description method is proposed. First

the rotating pixel difference convolution module is designed to symmetrically sample differential information from images. This module is further extended to implement the judger module that integrates sampled differential and convolutional information

effectively extracting features from the raw image. Finally

shared encoder and lightweight decoder

which coordinates information between decoders

are designed

while improving the loss function to construct an efficient encoder-decoder network model. Experimental results demonstrate that the proposed method maintains favorable real-time performance while achieving high computational accuracy

providing a solution that balances real-time capability and precision for applications such as image matching and visual localization.

关键词

Keywords

references

SCHONBERGER J L, FRAHM J M. Structure from motion revisited[C] //2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 4104-4113.

WANG J, ZHONG Y, DAI Y, et al . Deep two-view structure-from-motion revisited[C ] //2021 IEEE Conferenc e on Computer Vision and Pattern Recognition. Online: IEEE, 2021: 8953-8962.

XIU H, LIANG Y, ZENG H, et al . Robust self-supervised monocular visual odometry based on prediction-update pose estimation network[J ] . Engineering Applications of Artificial Intelligence, 2022, 116: 105481-105481.

MUR-ARTAL R, MONTIEL J M M, TARDOS J D. ORB-SLAM: a versatile and accurate monocular SLAM system[J]. Computing Research Repository, 2015, 31(5): 1147-1163.

CHEN W, SHANG G, JI A, et al . An overview on visual slam: from tradition to semantic[J ] . Remote Sensing, 2022, 14(13): 3010.

CHENG J, ZHANG L, CHEN Q, et al . A review of visual slam methods for autonomous driving vehicles[J ] . Engineering Applications of Artificial Intelligence, 2022, 114: 104992.

LOWE D. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-110.

DETONE D, MALISIEWICZ T, RABINOVICH A. Superpoint: self-supervised interest point detection and description[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) . Salt Lake City: IEEE, 2018: 224-236.

SUN J, SHEN Z, WANG Y, et al . LoFTR: detector-free local feature matching with transformers[C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Nashville: IEEE, 2021: 8918-8927.

LIU H, ZHAO L, PENG Z, et al . A low-cost and scalable framework to build large-scale localization benchmark for augmented reality[J ] . IEEE Transactions on Circuits and Systems for Video Technology, 2024, 34(4): 2274-2288.

WU S, OERLEMANS A, BAKKER E, et al . A comprehensive evaluation of local detectors and descriptors[J ] . Signal Processing: Image Communication, 2017, 59: 150-167.

GONG X, LIU Y, WU Q, et al . An accurate, robust visual odometry and detail-preserving reconstruction system[J ] . IEEE Transactions on Multimedia, 2021, 23: 2820-2832.

YE P C, LI D, ZHANG Y. Direct sparse visual odometer based on enhanced stereo-camera constraints[J]. Journal of Guangdong University of Technology, 2021, 38(4): 65-70.

ZHU F L, ZENG B, CAO J. Parallel optimization and implementation of SLAM algorithm[J]. Journal of Guangdong University of Technology, 2017, 34(2): 92-96.

JIANG X, MA J, XIAO G, et al . A review of multimodal image matching: methods and applications[J ] . Information Fusion, 2021, 73: 22-71.

MA J, JIANG X, FAN A, et al . Image matching from handcrafted to deep features: a survey[J ] . International Journal of Computer Vision, 2020, 129(1): 23-79.

XU S, CHEN S, XU R, et al . Local feature matching using deep learning: A survey[J ] . Information Fusion, 2024, 107: 102344.

WANG K, LENG C, YAN H, et al . Feature matching based on gaussian kernel convolution and minimum relative motion[J ] . Engineering Applications of Artificial Intelligence, 2024, 131: 107795.

BAY H, ESS A, TUYTELAARS T, et al . Speeded-up robust features (SURF) [J ] . Computer Vision and Image Understanding, 2008, 110(3): 346-359.

LEUTENEGGER S, CHLI M, SIEGWART R Y. BRISK: binary robust invariant scalable keypoints[C]//2011 International conference on computer vision. Barcelona: IEEE, 2011: 2548-2555.

ALCANTARILLA P F, BARTOLI A, DAVISON A J. KAZE features[C] //12th European Conference on Computer Vision (ECCV 2012) . Florence: Springer, 2012: 214-227.

ALCANTARILLA P F, SOLUTIONS T. Fast explicit diffusion for accelerated features in nonlinear scale spaces[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 34(7): 1281-1298.

ROSTEN E, PORTER R, DRUMMOND T. Faster and better: a machine learning approach to corner detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 32(1): 105-119.

RUBLEE E, RABAUD V, KONOLIGE K, et al . ORB: an efficient alternative to SIFT or SURF[C ] //2011 International Conference on Computer Vision. Barcelona: IEEE, 2011: 2564-2571.

YI K M, TRULLS E, LEPETIT V, et al . Lift: learned invariant feature transform[C ] //14th European Conference on Computer Vision (ECCV 2016) . Amsterdam: Springer, 2016: 467-483.

DUSMANU M, ROCCO I, PAJDLA T, et al . D2-Net: a trainable CNN for joint description and detection of local features[C ] //Proceedings of the IEEE/CVF Confe rence on Computer Vision and Pattern Recognition. Long Beach: 2019: 8092-8101.

LIN T Y, MAIRE M, BELONGIE S, et al . Microsoft coco: common objects in context[C ] //13th European Conference on Computer Vision (ECCV 2014) . Zurich: Springer, 2014: 740-755.

VASWANI A, SHAZEER N, PARMAR N, et al . Attention is all you need[J ] . Advances in Neural Information Processing Systems, 2017, 30: 5998-6008.

SU Z, LIU W, YU Z, et al . Pixel difference networks for efficient edge detection[C ] //2021 IEEE/CVF International Conference on Computer Vision (ICCV) . Montreal: IEEE/CVF, 2021: 5117-5127.

DING X, ZHANG X, MA N, et al . Repvgg: making vgg-style convnets great again[C ] //2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Nashville: IEEE/CVF, 2021: 13733-13742.

GEIGER A, LENZ P, URTASUN R. Are we ready for autonomous driving? the kitti vision benchmark suite[C] //2012 IEEE Conference on Computer Vision and Pattern Recognition. Rhode Island: IEEE, 2012: 3354-3361.

BALNTAS V, LENC K, VEDALDI A, et al . HPa tches: a benchmark and evaluation of handcrafted and learned local descriptors[C ] //2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Honolulu: IEEE, 2017: 5173-5182.

TYSZKIEWICZ M, FUA P, TRULLS E. DISK: learning local features with policy gradient[J]. Computing Research Repository, 2020, 33: 14254-14265.

SARLIN P E, DETONE D, MALISIEWICZ T, et al . Superglue: learning feature matching with graph neural networks[C ] //2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Seattle: IEEE/CVF, 2020: 4938-4947.

LINDENBERGER P, SARLIN P E, POLLEFEYS M. Lightglue: local feature matching at light speed[C] //2023 IEEE/CVF International Conference on Computer Vision (ICCV) . Paris: IEEE/CVF, 2023: 17627-17638.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于计算机视觉的腾空飞脚错误动作识别模型

具守恒量波动方程的循环双子网物理信息神经网络

基于高效时序建模的RGB-T跟踪

双模态迭代交叉注意力融合集成框架

基于数字混频的脉冲式激光测距方法的研究与实现