열화상 이미지 히스토그램의 가우시안 혼합 모델 근사를 통한 열화상-관성 센서 오도메트리
Infrared Visual Inertial Odometry via Gaussian Mixture Model Approximation of Thermal Image Histogram
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
We introduce a novel Visual Inertial Odometry (VIO) algorithm designed to improve the performance of thermal-inertial odometry. Thermal infrared image, though advantageous for feature extraction in low-light conditions, typically suffers from a high noise level and significant information loss during the 8-bit conversion. Our algorithm overcomes these limitations by approximating a 14-bit raw pixel histogram into a Gaussian mixture model. The conversion method effectively emphasizes image regions where texture for visual tracking is abundant while reduces unnecessary background information. We incorporate the robust learning-based feature extraction and matching methods, SuperPoint and SuperGlue, and zero velocity detection module to further reduce the uncertainty of visual odometry. Tested across various datasets, the proposed algorithm shows improved performance compared to other state-of-the-art VIO algorithms, paving the way for robust thermal-inertial odometry.
Abstract
We introduce a novel Visual Inertial Odometry (VIO) algorithm designed to improve the performance of thermal-inertial odometry. Thermal infrared image, though advantageous for feature extraction in low-light conditions, typically suffers from a high noise level and significant information loss during the 8-bit conversion. Our algorithm overcomes these limitations by approximating a 14-bit raw pixel histogram into a Gaussian mixture model. The conversion method effectively emphasizes image regions where texture for visual tracking is abundant while reduces unnecessary background information. We incorporate the robust learning-based feature extraction and matching methods, SuperPoint and SuperGlue, and zero velocity detection module to further reduce the uncertainty of visual odometry. Tested across various datasets, the proposed algorithm shows improved performance compared to other state-of-the-art VIO algorithms, paving the way for robust thermal-inertial odometry.
참고문헌 (32)
- Loic Landrieu, "Large-Scale Point Cloud Semantic Segmentation with Superpoint Graphs", 2018 IEEE, 2018, 4558~4567[DOI]
- Paul-Edouard Sarlin, "SuperGlue: Learning Feature Matching With Graph Neural Networks", IEEE Conference on Computer Vision and Pattern Recognition, pp.4938~4947 (2020), IEEE
- E. Rublee, "ORB : An efficient alternative to SIFT or SURF", IEEE International Conference on Computer Vision (ICCV), pp.2564~2571 (2011)
- David G. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints", International Journal of Computer Vision, 60(2), pp.91~110 (2004), Springer Science and Business Media LLC[DOI]
- Herbert Bay, "Speeded-Up Robust Features (SURF)", Computer Vision and Image Understanding, 110(3), pp.346~359 (2008), Elsevier BV[DOI]
- Paulo Vinicius Koerich Borges, "Practical Infrared Visual Odometry", IEEE Transactions on Intelligent Transportation Systems, 17(8), pp.2205~2213 (2016), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Stephen Vidas, "Hand-held monocular SLAM in thermal-infrared", 2012 12th International Conference on Control Automation Robotics & Vision (ICARCV), pp.859~864 (2012), IEEE
- "An optimization based approach to visual odometry using infrared images" (2010)[URL]
- Shehryar Khattak, "Keyframe‐based thermal–inertial odometry", Journal of Field Robotics, 37(4), pp.552~579 (2019), Wiley[DOI]
- Young-Sik Shin, "Sparse Depth Enhanced Direct Thermal-Infrared SLAM Beyond the Visible Spectrum", IEEE Robotics and Automation Letters, 4(3), pp.2918~2925 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Shibo Zhao, "TP-TIO: A Robust Thermal-Inertial Odometry with Deep ThermalPoint", 2020 IEEE, 2020, 4505~4512[DOI]
- Julien Poujol, "A Visible-Thermal Fusion Based Monocular Visual Odometry", Advances in Intelligent Systems and Computing, pp.517~528 (2015), Springer International Publishing[DOI]
- Tarek Mouats, "Multispectral Stereo Odometry", IEEE Transactions on Intelligent Transportation Systems, 16(3), pp.1210~1224 (2015), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Long Chen, "RGB-T SLAM: A flexible SLAM framework by combining appearance and thermal information", 2017 IEEE International Conference on Robotics and Automation (ICRA), pp.5682~5687 (2017), IEEE
- Shehryar Khattak, "Visual-Thermal Landmarks and Inertial Fusion for Navigation in Degraded Visual Environments", 2019 IEEE Aerospace Conference, pp.1~9 (2019), IEEE
- Tarek Mouats, "Thermal Stereo Odometry for UAVs", IEEE Sensors Journal, 15(11), pp.6335~6347 (2015), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Sen Wang, "DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks", 2017 IEEE International Conference on Robotics and Automation (ICRA), pp.2043~2050 (2017), IEEE
- Alex Kendall, "PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization", 2015 IEEE International Conference on Computer Vision (ICCV), pp.2938~2946 (2015), IEEE
- Iaroslav Melekhov, "Relative Camera Pose Estimation Using Convolutional Neural Networks", Advanced Concepts for Intelligent Vision Systems, pp.675~687 (2017), Springer International Publishing[DOI]
- D. DeTone, "Deep Image Homography Estimation", ~ (2016)
- Edgar Simo-Serra, "Discriminative Learning of Deep Convolutional Feature Point Descriptors", 2015 IEEE International Conference on Computer Vision (ICCV), pp.118~126 (2015), IEEE
- Kwang Moo Yi, "LIFT: Learned Invariant Feature Transform", European Conference on Computer Vision (ECCV), pp.467~483 (2016), Springer International Publishing
- Hyeonwoo Noh, "Large-Scale Image Retrieval with Attentive Deep Local Features", 2017 IEEE International Conference on Computer Vision (ICCV), pp.3456~3465 (2017), IEEE
- Dongjiang Li, "DXSLAM: A Robust and Efficient Visual SLAM System with Deep Features", 2020 IEEE, 2020, 4958~4965[DOI]
- Jiexiong Tang, "GCNv2: Efficient Correspondence Prediction for Real-Time SLAM", IEEE Robotics and Automation Letters, 4(4), pp.3505~3512 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Raul Mur-Artal, "ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras", IEEE Transactions on Robotics, 33(5), pp.1255~1262 (2017), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Huai-Jen Liang, "SalientDSO: Bringing Attention to Direct Sparse Odometry", IEEE Transactions on Automation Science and Engineering, 16(4), pp.1619~1626 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Jakob Engel, "Direct Sparse Odometry", IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(3), pp.611~625 (2018), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Anastasios I. Mourikis, "A Multi-State Constraint Kalman Filter for Vision-aided Inertial Navigation", Proceedings 2007 IEEE International Conference on Robotics and Automation, pp.3565~3572 (2007), IEEE
- Indirect Kalman Filter for 3D Attitude Estimation, "N. Trawny", University of Minnesota, ~, pp.Rep. 2005-002 (2005)
- Patrick Geneva, "OpenVINS: A Research Platform for Visual-Inertial Estimation", 2020 IEEE International Conference on Robotics and Automation (ICRA), pp.4666~4672 (2020), IEEE
- Tong Qin, "VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator", IEEE Transactions on Robotics, 34(4), pp.1004~1020 (2018), Institute of Electrical and Electronics Engineers (IEEE)[DOI]