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
Hanqiu found it when we evaluated and reproduced Floxels in the Argoverse 2 validation set locally.
# for local and online eval mask from official repo
s5cmd --no-sign-request cp "s3://argoverse/tasks/3d_scene_flow/zips/*".
The provided eval_mask at first 5-10 frames in each scenes will include ground points as evaluaiton and some flow annotation actually on the ground points.
Some discussions and images:
Maybe that's another reason behind why local validation set have lower performance than test set (leaderboard-v2 change eval_mask and eval frames so it didn't use the v1 one here). But it should be fair for all methods, since we set all ground points' relative flow=0 for all methods in opensf code. 😅
In this discussion thread, I will run all methods in argoverse 2 validation set and backup all methods in OpenSceneFlow so that we can easily get these correct results after masking out the ground points also.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Hanqiu found it when we evaluated and reproduced Floxels in the Argoverse 2 validation set locally.
The provided eval_mask at first 5-10 frames in each scenes will include ground points as evaluaiton and some flow annotation actually on the ground points.
Some discussions and images:

In this discussion thread, I will run all methods in argoverse 2 validation set and backup all methods in OpenSceneFlow so that we can easily get these correct results after masking out the ground points also.
The commit I ran is: Kin-Zhang/OpenSceneFlow@c5d2177. I will merge these changes in the next paper with fixed results here.
All reactions