Facebook contributes to MLPerf, open-sources Mask R-CNN2Go

Bill Jia

Facebook has long supported open standards and interoperability between frameworks and hardware for driving machine learning (ML) innovation with projects like Open Neural Network Exchange (ONNX) a…

Overview

Facebook has contributed to the MLPerf initiative by open-sourcing Mask R-CNN2Go, a computer vision model optimized for mobile and embedded devices. This effort aims to establish industry-standard benchmarks for machine learning performance across various applications.

What You'll Learn

1

How to utilize Mask R-CNN2Go for on-device machine learning applications

2

Why MLPerf benchmarks are essential for evaluating ML performance

3

When to apply edge inference benchmarks for mobile device optimization

Key Questions Answered

What is Mask R-CNN2Go and how is it used?

Mask R-CNN2Go is a computer vision model developed by Facebook, optimized for embedded and mobile devices. It supports various on-device ML use cases, including object detection, classification, person segmentation, and body pose estimation, enabling accurate real-time inference.

What role does Facebook play in the MLPerf initiative?

Facebook supports the MLPerf initiative by contributing to the development of industry-standard benchmarks for machine learning performance. This includes providing benchmark reference implementations for models like Mask R-CNN2Go and ShuffleNet, aimed at enhancing evaluation methods for ML applications.

How does Mask R-CNN2Go improve mobile ML applications?

Mask R-CNN2Go enhances mobile ML applications by providing a model specifically designed for mobile devices, enabling efficient on-device inference for tasks such as object detection and pose estimation. This optimization allows for real-time performance in applications like augmented reality.

What benchmarks does MLPerf provide?

MLPerf provides a comprehensive suite of benchmarks for measuring system-level performance of ML software frameworks, hardware accelerators, and cloud platforms. It covers diverse application use cases, including image classification, object detection, and speech-to-text translation, facilitating fair evaluations.

Technologies & Tools

Some links below are affiliate links. We may earn a commission if you make a purchase.

Machine Learning Model

Mask R-cnn2go

Optimized for embedded and mobile devices for tasks like object detection and pose estimation.

Machine Learning Framework

Caffe2

Current framework used to run Mask R-CNN2Go.

Machine Learning Framework

Pytorch

Future framework planned for Mask R-CNN2Go as it adds more capabilities.

Key Actionable Insights

1
Leverage Mask R-CNN2Go for developing real-time mobile applications that require object detection and segmentation.
This model is specifically optimized for mobile devices, making it suitable for applications in augmented reality and other interactive experiences.

2
Participate in the MLPerf initiative to contribute to the evolution of ML benchmarks.
Engaging with MLPerf can help you stay at the forefront of ML performance evaluation and ensure your applications meet industry standards.

3
Utilize the benchmarks provided by MLPerf to identify performance bottlenecks in your ML applications.
By understanding the benchmarks, you can optimize your systems for more efficient on-device inference, which is crucial as ML increasingly shifts to edge devices.