Closed ozyilmaz closed 2 years ago
DetectionOutput
is programmatic and requires looping, so there are no plans to implement it at this time.
Bug fixes https://github.com/PINTO0309/openvino2tensorflow/releases/tag/v1.27.4
Committed the transformed model without DetectionOutput. https://github.com/PINTO0309/PINTO_model_zoo/tree/main/227_face-detection-adas-0001
Reference demonstration program for post-processing https://github.com/PINTO0309/PINTO_model_zoo/tree/ce8421f2e8720d636d5f00e817debdc6720f235a/179_person-detection-0202/demo
Incidentally, this model, which can also detect backward-facing objects, is easier to handle. Since facial object detection and face orientation estimation are separated into separate models, only the face detection model can be changed to a lighter model. https://github.com/PINTO0309/HeadPoseEstimation-WHENet-yolov4-onnx-openvino https://github.com/PINTO0309/PINTO_model_zoo/tree/main/106_WHENet
Hello Katsuya,
First of all thanks much for your response. This helps a lot. One question: When we transform the model without DetectionOutput, how can we read the face detection outputs?
Regards, Ozgur
On 12 Jan 2022, at 03:09, Katsuya Hyodo @.***> wrote:
DetectionOutput is programmatic and requires looping, so there are no plans to implement it at this time.
Bug fixes https://github.com/PINTO0309/openvino2tensorflow/releases/tag/v1.27.4 https://github.com/PINTO0309/openvino2tensorflow/releases/tag/v1.27.4 Committed the transformed model without DetectionOutput. https://github.com/PINTO0309/PINTO_model_zoo/tree/main/227_face-detection-adas-0001 https://github.com/PINTO0309/PINTO_model_zoo/tree/main/227_face-detection-adas-0001 Reference demonstration program for post-processing https://github.com/PINTO0309/PINTO_model_zoo/tree/ce8421f2e8720d636d5f00e817debdc6720f235a/179_person-detection-0202/demo https://github.com/PINTO0309/PINTO_model_zoo/tree/ce8421f2e8720d636d5f00e817debdc6720f235a/179_person-detection-0202/demo Incidentally, this model, which can also detect backward-facing objects, is easier to handle. Since facial object detection and face orientation estimation are separated into separate models, only the face detection model can be changed to a lighter model. https://github.com/PINTO0309/HeadPoseEstimation-WHENet-yolov4-onnx-openvino https://github.com/PINTO0309/HeadPoseEstimation-WHENet-yolov4-onnx-openvino https://github.com/PINTO0309/PINTO_model_zoo/tree/main/106_WHENet https://github.com/PINTO0309/PINTO_model_zoo/tree/main/106_WHENet https://user-images.githubusercontent.com/33194443/149040755-2c064912-7d9e-40a2-bd2e-d1d2592c6aa3.gif — Reply to this email directly, view it on GitHub https://github.com/PINTO0309/openvino2tensorflow/issues/89#issuecomment-1010491478, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIMMM4LDD7XCSXEY44SP2OTUVTBDHANCNFSM5LWANJMA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub. You are receiving this because you authored the thread.
As I told you in my comment above, you can see it in the demo code. https://github.com/PINTO0309/PINTO_model_zoo/blob/main/179_person-detection-0202/demo/demo_person-detection-0202_onnx.py https://github.com/PINTO0309/PINTO_model_zoo/blob/main/179_person-detection-0202/demo/demo_person-detection-0202_tflite.py
Issue Type
Feature Request
OS
Ubuntu
OS architecture
x86_64
Programming Language
Other
Framework
OpenVINO
Download URL for ONNX / OpenVINO IR
https://download.01.org/opencv/2020/openvinotoolkit/2020.4/open_model_zoo/models_bin/3/face-detection-adas-0001/FP16/
Convert Script
openvino2tensorflow --model_path /home/user/workdir/openvino_models/face-detection-adas-0001.xml --output_float16_quant_tflite
Description
Hello,
First of all thanks for sharing the library. We are trying to convert the pretrained face detection model from Intel's repo. And having an error: "The DetectionOutput layer is not yet implemented." It would be great of you added this layer to the library.
Best
Relevant Log Output
Source code for simple inference testing code
No response