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
Using the Huggingface API with AutoProcessor, AutoModelForZeroShotObjectDetection, and post_process_grounded_object_detection() (following Grounded SAM 2), I receive some empty label strings in the results (results[0]["labels"]) or strings that include 2 classes. My classes are specified as
text=".".join([
"pot", "pan", ...
])
I have a few questions:
What are these labels and the boxes associated with them? Do the empty ones not meet the post processing threshold?
What is the common practice with dealing with them? Do we simply discard these boxes?
How are multiple classes assigned to the box?
Thank you very much!
The text was updated successfully, but these errors were encountered:
willjhliang
changed the title
Some Empty Labels
Some Empty Labels or Labels with Multiple Classes
Aug 11, 2024
Using the Huggingface API with
AutoProcessor
,AutoModelForZeroShotObjectDetection
, andpost_process_grounded_object_detection()
(following Grounded SAM 2), I receive some empty label strings in the results (results[0]["labels"]
) or strings that include 2 classes. My classes are specified asI have a few questions:
Thank you very much!
The text was updated successfully, but these errors were encountered: