search

facebookresearch / Mask2Former

Code release for "Masked-attention Mask Transformer for Universal Image Segmentation"
MIT License
2.59k stars 388 forks source link

Setting the config correct for numerous, small objects #191

Open s-weissbach opened 1 year ago

s-weissbach commented 1 year ago

Hello,

I need assistance in configuring the settings for a segmentation task that involves detecting multiple smaller objects. I've already modified the DETECTIONS_PER_IMAGE: 100 flag for inference, but I'm curious if there are specific parameters that can be adjusted during training to enable more detections or to fine-tune the expected size of the detected objects.

Any guidance on this matter would be greatly appreciated. Thank you!