I am using this model to detect objects in an NPU unit hardware, but the problem is the output of the model, is unable to get box and label class from the output.
Please answer the following questions for yourself before submitting an issue.
[Yes ] I am using the latest TensorFlow Model Garden release and TensorFlow 2.
[ Yes] I am reporting the issue to the correct repository. (Model Garden official or research directory)
[ Yes] I checked to make sure that this issue has not been filed already.
Prerequisites
I am using this model to detect objects in an NPU unit hardware, but the problem is the output of the model, is unable to get box and label class from the output. Please answer the following questions for yourself before submitting an issue.
1. The entire URL of the file you are using
https://tfhub.dev/iree/lite-model/ssd_mobilenet_v1_100_320/uint8/default/1
2. Describe the bug
The output of the model is not understandable the model with nms works fine with the code below. A clear and concise description of what the bug is.
The only model that works with my code is this one https://tfhub.dev/iree/lite-model/ssd_mobilenet_v1_100_320/uint8/nms/1
Since it is not optimised for int8 input and int8 output.
Please if someone can help me to understand how to get output points
box_predictor_output = interpreter.get_tensor(output_details[0]['index']) class_predictor_output = interpreter.get_tensor(output_details[1]['index'])
from this model https://tfhub.dev/iree/lite-model/ssd_mobilenet_v1_100_320/uint8/default/1