Business Results

  • Up to 24% faster real-time prediction

  • Up to 34% faster prediction in batch

  • Up to 154% faster real-time prediction after training optimization (quantization)

  • Up to 107% faster prediction in batch after training optimization (quantization)

  • Model footprint is reduced from 158 MB to 40 MB (up to 75% compressed)

author-image

By

View All Reference Kits

Background

It is crucial for businesses to understand their organization's existing processes and identify gaps and bottlenecks to improve productivity and user experiences. With the growing digitization of business processes, visual process discovery tools and techniques have found broad applications across multiple industries.

Visual process discovery (VPD):
 

  • Captures real-time interactions between users and workflows
  • Maps and analyses the workflows and provides objective data-driven insights to enhance processes
  • Identifies processes that can be automated
     

In the financial services industry, VPD helps improve and automate loan processing, suspicious activity report generation, customer onboarding, account opening and closure, know your customer (KYC), and other repetitive processes.

Solution

In collaboration with Accenture*, Intel developed a visual process discovery AI reference kit, which may assist you with building an optimal UI element detection model. This model can be used on website screen captures to test efficient and successful process automation (robotic process automation [RPA]) for your websites.

End-to-End Flow Using Intel® AI Software Products

 

This reference kit uses an advanced PyTorch*-based pretrained Faster R-CNN ResNet-50 model to perform transfer learning on a Roboflow* Website Screenshots dataset. This model is an object detection algorithm that enables the convolutional neural network (CNN) to learn the region proposals. The model is further trained to detect UI elements in the input website screen captures.

Edge-deployed AI models offer real-time analysis of UI element detection, conducting computation at extremely low latency and boosting the whole visual process discovery. Thus, by quantizing and compressing the model (from floating point to integer model), while maintaining a similar level of accuracy as the floating-point model, efficient use of underlying resources can be demonstrated when deployed on edge devices with low processing and memory capabilities.

This reference kit includes:
 

  • Training data
  • An open source, trained model
  • Libraries
  • User guides
  • Intel® AI software products

At a Glance

  • Industry: E-commerce, cross-industry
  • Task: Detect the UI elements (buttons, links, text, images, headings, fields, labels, IFrames) present in the website screen capture. Quantize FP32 model to int8
  • Dataset: 236 MB Roboflow dataset1
  • Type of Learning: Transfer learning
  • Models: Faster R-CNN ResNet-50
  • Output: UI elements (buttons, links, text, images, headings, fields, labels, IFrames)
  • Intel AI Software Products:
    • Intel® Extension for PyTorch* v1.13.0
    • Intel® Neural Compressor

 

Technology

Optimized with Intel AI Software Products for Better Performance

The Visual Process Discovery model is optimized with Intel Extension for PyTorch (v1.13.0). The pretrained Faster R-CNN ResNet-50 model is used and optimized to detect UI elements in the input website screen captures. Intel Neural Compressor is used to quantize the FP32 model to the int8 model.

Intel Extension for PyTorch and Intel Neural Compressor allow you to reuse your model development code with minimal code changes for training and inferencing.

Performance benchmark tests were run on Microsoft Azure* Standard_D8_v5 using 3rd generation Intel® Xeon® processors to optimize the solution.

Benefits

The Visual Process Discovery reference kit assists you with building an optimal UI element detection model that can be used on website screen captures to test efficient and successful process automation (Robotic Process Automation [RPA]) for your websites.

To build a webpage UI elements detection model for visual process discovery using the deep learning approach, machine learning developers may need to train models with a large dataset and run inference more frequently. The ability to accelerate training may allow them to train more frequently and achieve better accuracy. Faster inferencing speed may enable them to run prediction in real time as well as offline batch processing. The result is a better online customer experience.

With Intel® oneAPI components, little to no code change is required to attain the performance boost.

Download Kit

Reference

1. Website Screenshots Dataset, https://public.roboflow.com/object-detection/website-screenshots/1. See this dataset's applicable license for terms and conditions. Intel Corporation does not own the rights to this dataset and does not confer any rights to it.

Stay Up to Date on AI Workload Optimizations

Sign up to receive hand-curated technical articles, tutorials, developer tools, training opportunities, and more to help you accelerate and optimize your end-to-end AI and data science workflows.

Take a chance and subscribe. You can change your mind at any time.

By submitting this form, you are confirming you are an adult 18 years or older and you agree to share your personal information with Intel to use for this business request. Intel's web sites and communications are subject to our Privacy Notice and Terms of Use.
By submitting this form, you are confirming you are an adult 18 years or older and you agree to share your personal information with Intel to use for this business request. You also agree to subscribe to stay connected to the latest Intel technologies and industry trends by email and telephone. You may unsubscribe at any time. Intel's web sites and communications are subject to our Privacy Notice and Terms of Use.