1215232494

python 函数参数默认值如果是元组，元组元素逗号后的数值会被错误识别为参数 例如 def function(a, b=(1,2), c=3): 注释识别参数有四个：a、b、2、c

As mentioned in #833 This open source project “[segment-anything-annotator](https://github.com/haochenheheda/segment-anything-annotator)” has already implemented this function with the right mouse button. Can we integrate this function？

Mark lines for training but get areas

3

### Provide environment information Python 3.7.13 labelme 5.0.0 ### What OS are you using? Ubuntu 20.04.4 ### Describe the Bug I use labelme to label linestrip  and use [labelme2coco.py](https://github.com/wkentaro/labelme/blob/855615965168f0dfc3ec55e087dcdfe65b371eff/examples/instance_segmentation/labelme2coco.py#L1)...

Whether this function has been implemented at present?

Visualization of custom data results

1

How do I visualize results when testing custom data? When I use -- debug, an error will be reported: File "LSTR/db/custom.py", line 200, in __getitem__ img = cv2.imread(item['path']) KeyError: 'path'

I trained a model using base_coco.yaml files,but when I want to visualize the detection results of Coco data images instead of videos, I encountered the same problem as [#5 ](https://github.com/sukjunhwang/IFC/issues/5#issuecomment-1236831096)...