Hair Segmentation Using Heuristically-Trained Neural Networks.

摘要

We present a method for binary classification using neural networks that performs training and classification on the same data using the help of a pre-training heuristic classifier. The heuristic classifier is initially used to segment data into three clusters of high confidence positives, high confidence negatives, and low confidence sets. The high confidence sets are used to train a neural network (NN) which is then used to classify the low confidence set. Applying this method to the binary classification of hair vs. non-hair patches, we obtain a 9% performance increase using the heuristically-trained NN over the current state of the art hair segmentation method.