FastSAM

FastSAM copied to clipboard

Do you have plan to publish your training code?

2

Hi, Thank you for your good work. May I ask, do you have plan to publish your training code?

rbli-john

This code can not put masks for specific color correctly, for example, use your code on colab with text prompt "blue area" on that "dog.jpg" the mask will be on...

BWDforce

How can I get the size of each segmented area in the "everything prompt" inference result?

JooHyungG

hello all, we are trying to cog build the docker image but failed: ``` cog build --no-cache Building Docker image from environment in cog.yaml as cog-fastsam... [+] Building 451.5s (14/21)...

timiil

as the title, how to use in GPU mode? thanks very much

timiil

 我的text prompt页面不能更换下方的示例图片，只能停留在dog 

alps008

I am moving from Facebook SAM to FastSAM for performance This is the output from Facebook SAM which is a one-dimension Tensor object, a mask ```python tensor([[[[ 0.0513, 0.0712, 0.0959,...

hyfung

Hi, I was plotting the bounding boxes on masks for visualiztion. boxes = everything_results[0].boxes.data and masks = everything_results[0].boxes.data # converted xywh format to xyxy format. Bounding boxes are not surrounding...

qevr

Contour Line Fitting

1

Hi there, I notice the contour line of the segmetation is sometimes not accurate, like the figure shows, it shall be a straight line instead of a wavy line. Do...

NTUZZH

args = parse_args() print("model loaded 1 args.device, ", args.device) # 使用全局变量中的模型实例进行推理 everything_results = model(input_image, device=args.device, retina_masks=args.retina, imgsz=args.imgsz, conf=args.conf, iou=args.iou) print("model loaded 2 " , everything_results) ----- everything_results = model(...) 这一步遇到一些大图片时会卡住，外部接口调用报错，但是内部程序无任何反馈，一直卡住没反应...

alexhmyang