OpenVino DL Workbench Optimization dataset(WIDERFACE)

[AI_Jun]

Thank you Intel for providing OpenVino Toolkit!

I'm using DL WorkBench.
In particular, I am using an optimization tool, and I am curious about the Dataset.

I want to optimize the file for face detection using the widerface dataset. It must be annotated to verify accuracy be annotated.

Is it possible to check the accuracy like coco dataset and imagenet dataset if we annotate the widerface dataset like sample dataset ?

[Iffa_Intel]

Greetings,

Fyi,

• Model Downloader to download models from the Intel® Open Model Zoo with pretrained models for a range of different tasks
• Model Optimizer to transform models into the Intermediate Representation (IR) format
• Post-Training Optimization toolkit to calibrate a model and then execute it in the INT8 precision
• Accuracy Checker to determine the accuracy of a model
• Benchmark Tool to estimate deep learning inference performance on supported devices

Here is the official documentation:

https://docs.openvinotoolkit.org/latest/workbench_docs_Workbench_DG_Introduction.html

Sincerely,

Iffa

[Iffa_Intel]

Greetings,

Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question. 

Sincerely,

Iffa

[Alexander_D_Intel1]

Hello @AI_Jun !

Thank you for asking. Currently, WIDERFACE dataset is not supported. You can find the list of datasets supported in the DL Workbench: https://docs.openvinotoolkit.org/latest/workbench_docs_Workbench_DG_Dataset_Types.html

If you have any additional questions or proposals, please ask them at the official DL Workbench Discussion Forum: https://github.com/openvinotoolkit/workbench_feedback/discussions. For example, you can request support of WIDERFACE dataset there.