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
Thank you for providing the project code. During the testing process, I found that the predicted metric depth result was not ideal. Is this due to different camera parameters? If so, which part of the project can adjust relevant parameters to achieve more ideal testing results?
The text was updated successfully, but these errors were encountered:
What would you mean by "non-ideal", the scale is off or is ti missing details?
Remember that for unbounded scenes (like aerial or outdoor) the scene could be off especially for far away objects.
I am testing an outdoor farmland image, and in the predicted depth image, the GT between the farmland and the camera is 10m, but the predicted value is 23m.
Thank you for providing the project code. During the testing process, I found that the predicted metric depth result was not ideal. Is this due to different camera parameters? If so, which part of the project can adjust relevant parameters to achieve more ideal testing results?
The text was updated successfully, but these errors were encountered: