'Compute distance and position from camera (lat,lon) to object on photo
I just started working on a new project, doing research about the state of the art methods computing the position of objects on photos, given the following info:
Input data:
- Camera position (lat,lon) where the photo was taken
- Camera parameters (heading, sensor/lens data)
- Bounding box of ROI (annotated objects on img)
- Sequence of consecutive photos
Desired output:
- Position of ROI (lat, lon), ideally with an error < 1m
- Distance from ROI to camera (Haversine?)
How could I achieve such a performance? Any help would be much appreciated! Thanks, Tony
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|