LaPa Face Parsing Dataset

LaPa stands for Landmark guided face Parsing, which is an image dataset for face parsing. The dataset contains 22,176 facial images with various expressions, postures, and occlusions. Each image has 11 types of pixel-level annotations and 106 facial key points. The training set has 18,176 images, the validation set has 2,000 images, and the test set has 2,000 images.