How to Implement the Dual Shot Face Detector (DSFD) Demo on MacOS:

Original article was published by The Altruist on Deep Learning on Medium


New Programmer’s Guide:

How to Implement the Dual Shot Face Detector (DSFD) Demo on MacOS:

A complete step-by-step tutorial from start to finish

Acknowledgments:

@oyerst solved this issue

@hypadr1v3 solved this issue

@lijiannuist solved this issue and that issue

@vlad3996 solved this issue

@yihongXU solved this issue

Introduction:

This article is for those of us still early in our programming path — keep going

If you’re still new to programming, I encourage you to complete this tutorial because DSFD is worth it and going through this process will stretch you.

It’s intentionally thorough, so you should have everything you need.

But, if something isn’t covered, leave a comment and we’ll figure it out.

Background:

In this paper, Tencent introduces its open-sourced Dual Shot Face Detector (DSFD) algorithm. DSFD is a new face detection network that addresses three key areas of facial detection. This includes better feature learning, progressive loss design, and anchor assign based data augmentation.

In 2018, the DSFD algorithm ranked first across the board in the WIDER FACE Face Detection Benchmark.

The benchmark evaluation tested the detection systems for mean average precision (mAP) on set portions of the dataset split into three difficulty levels.

Validation set scores:

  • 96.6% (mAP) in the easy difficulty level
  • 95.7% (mAP) in the medium difficulty level
  • 90.4% (mAP) in the hard difficulty level

Test set scores:

  • 96.0% (mAP) in the easy difficulty level
  • 95.3% (mAP) in the medium difficulty level
  • 90.0% (mAP) in the hard difficulty level