Skip to main content

QC on Optimization Results

info

This feature is still work in progress. Expect more updates here in the near term.

In this tutorial, we will guide you through the process of using RapidPipeline's new feature for quantitative evaluation of your output models. This feature provides you with an evaluation score and a detailed HTML report for any newly optimized model. Let's get started!

Viewing the Evaluation Score​

RapidPipeline QC

  1. Access the Sidebar: After the optimization process is complete, navigate to the sidebar in RapidPipeline's interface.
  2. Find the Evaluation Score: In the sidebar, you will see the evaluation score for your newly optimized model. This score provides a quantitative measure of the output model's quality based on its visual similarity to the input model.

Reviewing the HTML Report​

  1. Alongside the evaluation score in the sidebar, you will find a link to an HTML report. This will initiate a download.
  2. Once you unpack the downlaod contents you will find the report inside a path like QCTools >> Report >> qa-page-...html
    • There are other files alongside the report here, mainly the image renders of the original and each export format.
  3. When you open the report you will see three sections:
    • Original Scene: These are renders of the input file
    • Comparison Scene: These are renders of the exported format
    • Pixel Difference: These are pixel comparisons between the two

If you click on any of the images in Original Scene or Comparison Scene, you will have a slider that allows you to evaluate the differences easier:

RapidPipeline QC Report

Future Improvements​

  1. The HTML report and image comparison will eventually be integrated into the rest of the UI.
  2. The score evaluation for optimization is still being improved.
  3. Sometimes there can be a slight delay after 3D Processor has finished running and QC results appearing in the sidebar.

Conclusion​

With RapidPipeline's new quantitative evaluation feature, you can now objectively assess the quality of your optimized models. By following this tutorial, you can easily find and interpret the evaluation score and detailed HTML report for any newly optimized model. Stay tuned for future updates as we continue to improve this feature.