Display bounding box on st.image based on bounding box

Hi!

I’m new to the Streamlit community and I’m working on a computer vision app. I’m wondering if there is the functionality within Streamlit to display or hide bounding box detections on images based on a checkbox. Basically I want the app to function like this:



The difference in the real app would be that around 12 images are displayed at one time. Any advice on how I might implement this would be greatly appreciated!

Are you already able to display the image with the bounding boxes?

The boxes in the image above I drew in MS Paint to show how I want it to look. I was able to get it to display using patches in matplotlib and st.pyplot(), but ideally I wouldn’t have to put images as the background of a plot in order to do this.

Streamlit doesn’t provide such functionality. It will display your images, but you need other tools to create the images themselves. You can certainly use a plotting library and matplotlib is probably the best for this. I think opencv can do it too, but I have never used it so I cannot give you the details. Sorry, not really my area of expertise.

1 Like

No worries I appreciate the help, I ended up using opencv and it’s working well enough for what I need!

1 Like