TP-TIO: A Robust Thermal-Inertial Odometry with Deep ThermalPoint

To achieve robust motion estimation in visually degraded environments, thermal odometry has been an attraction in the robotics community. However, most thermal odometry methods are purely based on classical feature extractors, which is difficult to establish robust correspondences in successive frames due to sudden photometric changes and large thermal noise. To solve this problem, we propose ThermalPoint, a lightweight feature detection network specifically tailored for producing keypoints on thermal images, providing notable anti-noise improvements compared with other state-of-the-art methods. After that, we combine ThermalPoint with a novel radiometric feature tracking method, which directly makes use of full radiometric data and establishes reliable correspondences between sequential frames. Finally, taking advantage of an optimization-based visual-inertial framework, a deep feature-based thermal-inertial odometry (TP-TIO) framework is proposed and evaluated thoroughly in various visually degraded environments. Experiments show that our method outperforms state-of-the-art visual and laser odometry methods in smoke-filled environments and achieves competitive accuracy in normal environments.

[1]  Ji Zhang,et al.  LOAM: Lidar Odometry and Mapping in Real-time , 2014, Robotics: Science and Systems.

[2]  Vincent Lepetit,et al.  LIFT: Learned Invariant Feature Transform , 2016, ECCV.

[3]  Ayoung Kim,et al.  Sparse Depth Enhanced Direct Thermal-Infrared SLAM Beyond the Visible Spectrum , 2019, IEEE Robotics and Automation Letters.

[4]  Tomasz Malisiewicz,et al.  SuperPoint: Self-Supervised Interest Point Detection and Description , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[5]  John Folkesson,et al.  GCNv2: Efficient Correspondence Prediction for Real-Time SLAM , 2019, IEEE Robotics and Automation Letters.

[6]  Yasin Almalioglu,et al.  DeepTIO: A Deep Thermal-Inertial Odometry With Visual Hallucination , 2020, IEEE Robotics and Automation Letters.

[7]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.

[8]  Christos Papachristos,et al.  Keyframe‐based thermal–inertial odometry , 2019, J. Field Robotics.

[9]  Guoquan Huang,et al.  Robocentric visual–inertial odometry , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[10]  Angel Domingo Sappa,et al.  A Visible-Thermal Fusion Based Monocular Visual Odometry , 2015, ROBOT.

[11]  Simon Baker,et al.  Equivalence and efficiency of image alignment algorithms , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[12]  Nabil Aouf,et al.  Multispectral Stereo Odometry , 2015, IEEE Transactions on Intelligent Transportation Systems.

[13]  Gary R. Bradski,et al.  ORB: An efficient alternative to SIFT or SURF , 2011, 2011 International Conference on Computer Vision.

[14]  Daniel Cremers,et al.  Direct Sparse Odometry , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Michael Bosse,et al.  Keyframe-based visual–inertial odometry using nonlinear optimization , 2015, Int. J. Robotics Res..

[16]  Wolfram Burgard,et al.  A benchmark for the evaluation of RGB-D SLAM systems , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[17]  Bo Chen,et al.  MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications , 2017, ArXiv.

[18]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[19]  Shaojie Shen,et al.  VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator , 2017, IEEE Transactions on Robotics.

[20]  Paulo Vinicius Koerich Borges,et al.  Practical Infrared Visual Odometry , 2016, IEEE Transactions on Intelligent Transportation Systems.

[21]  Carlo Tomasi,et al.  Good features to track , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[22]  Naira Hovakimyan,et al.  Multi-Spectral Visual Odometry without Explicit Stereo Matching , 2019, 2019 International Conference on 3D Vision (3DV).