'How to GEOLOCATE an object of an image of street view?
I have developed a YoLo based algorithm in Python to detect objects from Google Maps images. Now, i have the task to GEOLOCATE that object in the image. I have tried to access the EXIF information of the image and adapt it to the object, but it gets high position errors.
Is there any way or technique to acces geolocation of an object detected of my image?
Solution 1:[1]
Unless Google provides an API to map Street View image pixels to vectors in North-East-Down coordinates, you can forget about doing it in general.
Street View images are the result of a very complex image processing pipeline, whose result is a cubic panorama. In the final web client view you have access to the location of the view (as latitude, longitude and elevation), and the horizontal direction of view (as an angle with respect to North), but that's about it. In order to geolocate an object in the scene you'd need to know its distance from the camera and the direction under which you see it.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Francesco Callari |