Instance segmentation output misses objects further away #393
AniketDhar
started this conversation in
General
Replies: 1 comment 1 reply
-
|
No label slacked |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
What does it mean? |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I am trying out instance segmentation with GSA to identify traffic poles from a scene. The examples I tried manage to detect traffic poles that are close but totally misses the ones in the back. Is there any reason for this? Can I fix this with some configurable parameters?
python grounded_sam_demo.py --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py --grounded_checkpoint groundingdino_swint_ogc.pth --sam_checkpoint sam_vit_h_4b8939.pth --input_image assets/german_road_2.jpg --output_dir "outputs" --box_threshold 0.25 --text_threshold 0.2 --text_prompt "traffic pole" --device "cpu"Beta Was this translation helpful? Give feedback.
All reactions