Only object detection

[Protozzzer]

Can I run mask rcnn or cascade mask rcnn ConvNeXt training with a custom dataset that has no segmentation images, only markup with object detection? I am getting an error related to masks on startup even when setting the with_mask=False flag in the config files -> "KeyError: 'qt_masks' ".

[FRunyang]

Hello, have you addressed this problem? During inference, how can I get only the bbox results?

Can I run mask rcnn or cascade mask rcnn ConvNeXt training with a custom dataset that has no segmentation images, only markup with object detection? I am getting an error related to masks on startup even when setting the with_mask=False flag in the config files -> "KeyError: 'qt_masks' ".

[yanjin340121]

SwinTransformer/Swin-Transformer-Object-Detection#25