output bounding boxes about imageai HOT 5 CLOSED

Thank you for bringing this up. We will provide options to obtain the bounding box details of each object detected in the next version of ImageAI. It will be announced soon. Keep in touch.

from imageai.

You can run object detection on any image parsed into the detectObjectsFromImage() function. Are you asking to obtain the x, y, width and height values for the bounding box of each object detected?

from imageai.

@OlafenwaMoses thank you for your reply. Yes, I just want to get the objects bounding boxes values (x, y, width, height) from images (not from the coco dataset). I have tired to use detections[0, :, :4] in the function detectObjectsFromImage() to output bboxes, the output is with shape (160965,4). It is difficult to acquire the expected values from the output without filtering through the preset minimum_percentage_probability and the calculated (label, score) by the model. Could you give me a cue to resolve this problem? Thank you again!!

from imageai.

The latest version of ImageAI (v2.0.2) is now available and supports retrieving x1,y1,x2 and y2 coordinates of each detected object. Find links to documentation below.

http://imageai.readthedocs.io/en/latest/detection/index.html

http://imageai.readthedocs.io/en/latest/video/index.html

from imageai.

Thank you for your work. I have installed the new version yesterday, it works well. Thank you again!!

from imageai.