What is the exact use case of Minimum face size in ML KIT?

Question

As per the official documentation,

Minimum Face Size: float (default: 0.1f) - The minimum size, relative to the image, of faces to detect.

So, is it like it will detect the faces only if it's matches this size ?

I've tested and couldn't get any reasonable results.

I don't want the face to be detected if it's too small.

Hi Rajesh, can you please elaborate on what you've tried? What was your expectation and what was the outcome? — Samuel Bushi, May 21 '20 at 18:33
Hi Rajesh, have you found any solution for this? I got stucked with same problem. — Selmeny, Aug 10 '20 at 03:17

score 2 · Answer 1 · answered Jun 08 '20 at 21:11

2

https://developers.google.com/android/reference/com/google/mlkit/vision/face/FaceDetectorOptions.Builder#setMinFaceSize(float)

Based on the documentation, the Face size is the minimum size to be detected. If you do not want to detect too small faces, please increase the number, i.e, setting it as 0.5 or 0.8 to only detect large faces.

answered Jun 08 '20 at 21:11

Shiyu

875
4
5

2

This doesn't work as written in docs. Even if you set minFaceSize to 1, it detects small faces – Devansh Maurya Aug 25 '20 at 10:05
Having the same issue. The parameter doesn't seem to have an effect on the min size of the face detected. – Peter Chaula Jan 08 '21 at 09:05

What is the exact use case of Minimum face size in ML KIT?

1 Answers1