-
Notifications
You must be signed in to change notification settings - Fork 1.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Support downsample in dataset #973
Support downsample in dataset #973
Conversation
@@ -266,6 +266,10 @@ class VanillaDataManagerConfig(InstantiateConfig): | |||
camera_optimizer: CameraOptimizerConfig = CameraOptimizerConfig() | |||
"""Specifies the camera pose optimizer used during training. Helpful if poses are noisy, such as for data from | |||
Record3D.""" | |||
dataset_camera_scale_factor: float = 1.0 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Maybe camera_res_scale_factor
. I worry this may be ambiguous with extrinsic scale.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks fine to me. I think we should still suggest that people pre-scale their images. We actually used to have this feature and found that people would upload 10mp images and everything would grind to a halt :)
Fair enough. Agree - it is probably a judgement call on which of the above methods are appropriate. Let me know if you agree that this feature in additional also makes sense |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, I think having the option is generally a good thing.
* init * tests passing * pass through arguments * verbose * remove debug prints * switch to bilinear * update viewer * make parser outputs private * change name
As discussed briefly in #957 - it seems like have downsample support in the base InputDataset could be beneficial. While colmap outputs images at multiple-resolutions this may not be the case for many other applications where there are images only stored at a single resolution. This allows us to load in data in the raw format and downsample once loaded to disk