Fig. 1. Perspective projection imaging model of a camera

Fig. 2. Position and orientation monocular visual measurement model

Fig. 3. 3D reconstruction model of stereo visual measurement system

Fig. 4. Position and orientation measurement based on 3D reconstruction of stereo visual measurement system

Fig. 5. Development of classical object pose measurement methods based on point feature

Fig. 6. Comparesion of different PnP algorithms ^[16]. (a) Ordinary-3D point configuration; (b) quasi-singular point configuration; (c) planar point configuration

Fig. 7. Development of deep learning object pose measurement methods based on point feature

Fig. 8. Deep learning based visual pose measurement using point feature. (a) PVNet ^[22] based on sparse keypoints; (b) CDPN ^[23] based on dense points; (c) GDR-Net ^[27] based on points-guided direct regression

Fig. 9. Space geometry diagram of the PnL problem^[34]

Fig. 10. Development of PnL methods

Fig. 11. Process of pose iterative method based on contour features ^[45]. (a) RGB image; (b) extracted contour features; (c) 2D contour projections corresponding to initial pose (red) and optimized pose (green); (d) model projection corresponding to optimized pose

Fig. 12. Development of pose iterative methods based on edge and contour feature

Fig. 13. Time line of region-based pose measurement method

Fig. 14. Local area correspondence of target surface^[50]

Fig. 15. Estimation of foreground and background probability based on different shape of areas. (a) Local circular area^[57]; (b) banded area^[61]

Fig. 16. Development of object pose measurement methods based on multiple features

Fig. 17. Object pose measurement results under strong occlusion ^[65]

Fig. 18. Pose measurement method under hybrid representations ^[68]

Fig. 19. Aircraft pose estimation approach with keypoints and structures^[21]

Fig. 20. Development of object pose measurement methods based on 3D reconstruction

Fig. 21. Multi-view pose measurement methods. (a) Multi-view 3D keypoints estimation^[71]; (b) multi-view 3D bounding box estimation^[72]

Fig. 22. Disparities estimation method. (a) Semiglobal matching (SGM) stereo method^[74]; (b) GC-Net: geometry and context for deep stereo regression^[75]

Fig. 23. Multi-source fusion pose measurement algorithm based on vision/IMU ^[83]

Fig. 24. Multi-source fusion pose measurement algorithm based on GPS/vision ^[87]

Fig. 25. Multi-source fusion pose measurement algorithm based on IMU/GPS/vision/UWB^[89]

Fig. 26. FFB6D algorithm framework using RGB-D image as input ^[90]

Fig. 27. Coordinate system setting in simulation experiment^[94]

Fig. 28. Model of aircraft landing process^[95]

Fig. 29. Experiment of aircraft landing guidance system^[97]. (a) Schematic diagram of the experiment; (b) experimental setup

Fig. 30. Trinocular visual system of Madrid University of Technology ^[99]

Fig. 31. Stereo vision guidance system in the test of UAV autonomous landing^[100]

Fig. 32. Visual measurement and recovery system of Portuguese Naval ship-based UAV ^[101]

Fig. 33. Monocular pose measurement system applied to aircraft test task^[68] .(a) Monocular pose measurement system model; (b) real-time pose results

Fig. 34. SMART-OLEV docking with the client satellite^[103]

Fig. 35. Visual measurement of muzzle vibration^[104]

Fig. 36. Image sequence of embedded missile separation under different attack angles of carrier aircraft^[105]. (a) Attack angles is 0°; (b) attack angle is 2°; (c) attack angle is 3°

Fig. 37. Test of projectile penetrating ice target ^[106]. (a) Imaging with normal high speed camera; (b) imaging with high speed infrared camera

Fig. 38. High speed camera and position and attitude interpretation ^[107]. (a) Before hitting the target; (b) the hitting moment; (c) 0.015 s after; (d) 0.0875 s after

Fig. 39. Aircraft cabin assembly in airbus assembly line^[108]

Fig. 40. TriDAR on board Space Shuttle Discovery^[109]

Fig. 41. Cabin automatic docking platform^[111]

Fig. 42. Automatic assembly equipment for low pressure turbine shaft of aero-engine^[112]

Fig. 43. High precision micro-vision automatic assembly system^[113]

Fig. 44. Human action recognition^[123]

Fig. 45. Human pose tracking^[126]

Fig. 46. Human pose estimation in the 2022 Beijing Winter Olympics. (a) 3DAT athlete tracking system^[130]; (b) athlete scoring system^[131]

Fig. 47. Pilot pose estimation^[134]