vedoua
You want to run inference on a pre-recorded video? You can do that by sending the frames to the device one by one.
hostInQ = device.getInputQueue("host_in", maxSize=4, blocking=False)
frame = cv2.resize(frame, (800, 800)) # resize your video frame
h, w, c = frame.shape
bgr_planar = frame.transpose(2, 0, 1).flatten()
imgFrame = dai.ImgFrame()
imgFrame.setType(dai.ImgFrame.Type.BGR888p)
imgFrame.setWidth(w)
imgFrame.setHeight(h)
imgFrame.setData(bgr_planar)
hostInQ.send(imgFrame)