different result

[sulaiman1988]

when applying the code below the result is equal to Tag2Text not equal to the RAM result as you refer to in the repo. export CUDA_VISIBLE_DEVICES=0 python automatic_label_ram_demo.py \ --config GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py \ --ram_checkpoint ./Tag2Text/ram_swin_large_14m.pth \ --grounded_checkpoint groundingdino_swint_ogc.pth \ --sam_checkpoint sam_vit_h_4b8939.pth \ --input_image assets/demo9.jpg \ --output_dir "outputs" \ --box_threshold 0.25 \ --text_threshold 0.2 \ --iou_threshold 0.5 \ --device "cuda"

[rentainhe]

Hello! Would you like to provide the visualization of your inference results? So we can compare the results of the official demo in Grounded-SAM

[wpl427]

I suspect it was done by hand