A CNN architecture that employs multi-task learning to detect the presence of a face mask and its coverage of the nose, mouth, and chin on human faces. Conducted training on MaskedFace-Net 138k synthetic images.