I have just started to get acquainted with teachablemachine. I used their website to create a simple object classification model.
No problem, I’m getting the right results! But I’m wondering how do I get the coordinates of the detected object? Any ideas?
Hi @push_ca,To get the bounding boxes from the predictions, you have to extract the sub list ['‘detection_boxes’] from the predictions which contain the bounding box co-ordinates. For example, i have an object detection model called detector
detector = hub.load(module_handle).signatures['default']
For making predictions i use
result = detector(converted_img)
To get the bounding boxes from results
result['detection_boxes']
Please refer to this gist for working code example. Thank You.