I have experimented with a pipeline to run the fast_depth.hef pretrained model from hailo model zoo. However, I have not been able to find much information on the return values after inputting into the model.
pipeline_str = (
"libcamerasrc ! "
"video/x-raw,format=NV12,width=1280,height=720,framerate=30/1 ! "
"videoscale ! videoconvert ! "
"video/x-raw,format=RGB,width=224,height=224 ! "
f"hailonet hef-path={self.hef_path} ! "
"videoconvert ! video/x-raw,format=NV12 ! "
"appsink name=mysink"
)
However, after the inference, I run this code:
roi = hailo.get_roi_from_buffer(buffer)
depth_tensors = roi.get_objects_typed(hailo.HAILO_DEPTH_MASK)
but depth_tensors seems to have nothing. I’m not sure what HAILO_DEPTH_MASK returns. Could anyone guide me to documentation which may help me with understanding the models on hailo model zoo, and how to implement a successful inference pipeline?