You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I sometimes take pictures with different FOV, e.g. one with my wide-sensor, and one with the main sensor. Or even a panorama or 360° picture. In consequence those have the same location, and the same angle, despite having different information.
There should be an option to consider those pictures not to be duplicates, as they may contain vastly different information.
Not sure if there should be a separate value for vertical and horizontal, e.g. to distinguish 360° image spheres from panoramas, which may span 360° in horizontal direction, but may be limited vertically.
The text was updated successfully, but these errors were encountered:
You can specify what angle difference to be considered as duplicate
--duplicate_angle DUPLICATE_ANGLE
The maximum camera angle difference between two images to be considered as heading in the same direction. If both images are also close to each other (see --duplicate_distance), the
later image will be marked as duplicate and will not be upload. [default: 5.0]
If your image FoV are, for example, 180 degrees, then specify --duplicate_angle=180. Does it work?
This would filter regarding heading, not Field of View. That is exactly what I am talking about. Those images would have (almost the same) heading, so regardless how small the angle is, they would be delete. Even if one shows a broad overall, but low res 180 degree photo and the other a very detailed zoomed in image.
I sometimes take pictures with different FOV, e.g. one with my wide-sensor, and one with the main sensor. Or even a panorama or 360° picture. In consequence those have the same location, and the same angle, despite having different information.
There should be an option to consider those pictures not to be duplicates, as they may contain vastly different information.
Not sure if there should be a separate value for vertical and horizontal, e.g. to distinguish 360° image spheres from panoramas, which may span 360° in horizontal direction, but may be limited vertically.
The text was updated successfully, but these errors were encountered: