apply non-maxima suppression to suppress weak, overlapping bounding boxes

Hi all,

i want to use the opencv.dnn function to filter overlapping bounding boxes:

# apply non-maxima suppression to suppress weak, overlapping
idxs = cv2.dnn.NMSBoxes(boxes, confidences, args["confidence"], args["threshold"])

any ideas to run this on the nano/xavier, without building opencv from source.

best regards martin

Hi m.fischer,

I can’r answer that due to no experience on opencv.dnn, may other developer share experiences or have suggestions with you.