MTCNN Face Detection v1.2.4

• Fixes bug when using a GPU in R2020b (#20)

MTCNN Face Detection v1.2.3

• Fixes bug in relative ordering of boxes and landmarks as well as potential error when there are overlapping boxes (#17)
• 2 point version increase to be able to use GitHub releases integration on File Exchange.

MTCNN Face Detection v1.2.1

• Fixes bug when using UseGPU=True with UseDagNet=True (i.e. R2019a support) (#12)

MTCNN Face Detection v1.2.0

• Adds support for R2019a
  • Uses DAGNetwork versions of the CNNs for compatibility with R2019a
• detect and detectFaces functions accept uint8, single, and double image types
  • singles and doubles are expected to be in the range 0-1
  • uint8s are expected to be in the range 0-255
• fixes a bug in proposeRegions for certain image sizes.

MTCNN Face Detection v1.0.1

• Adds support for Name Value pair optional arguments to mtcnn.detectFaces

MTCNN Face Detection v1.0

Version 1.0 allows face detection and alignment using a pre-trained MTCNN detector.

To install download the .mltbx file and open using MATLAB.

Requirements:

• MATLAB R2019b or later
• Deep Learning Toolbox
• Computer Vision Toolbox
• Image Processing Toolbox