Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks
share
We present a new loss function, namely Wing loss, for robust facial landmark localisation with Convolutional Neural Networks (CNNs). We first compare and analyse different objective functions and show that the L1 and smooth L1 loss functions perform much better than the widely used L2 loss function in facial landmark localisation. The analysis of these loss functions suggests that, for the training of a CNN-based localisation model, more attention should be paid to small and medium range errors. To this end, we design a piece-wise loss function. The new loss function amplifies the impact of errors from the interval (-w,w) by switching from L1 loss to a modified logarithm function. To address the problem of under-representation of samples with large out-of-plane head rotations in the training set, we propose a simple but effective boosting strategy, referred to as Hard Sample Mining (HSM). In particular, we deal with the data imbalance problem by duplicating the minority training samples and perturbing them by injecting random image rotation, bounding box translation and other data augmentation approaches. Last, the proposed approach is extended to create a two-stage localisation framework for robust facial landmark localisation in the wild. The experimental results obtained on the AFLW and 300W datasets demonstrate the merits of the Wing loss function, and prove the superiority of the proposed method over the state-of-the-art approaches.
READ FULL TEXT
