Hierarchical facial landmark localization via cascaded random binary patterns

• A regression framework is designed for multi-view facial landmark localization.
• The use of comparison based feature is highly efficient for landmark localization.
• Gradient-boosted decision tree is superior to random forest for localization task.
• Accuracy and speed are tested on the widely used open dataset: LFW, AFLW and 300-W.

The main challenge of facial landmark localization in real-world application is that the large changes of head pose and facial expressions cause substantial image appearance variations. To avoid high dimensional facial shape regression, we propose a hierarchical pose regression approach, estimating the head rotation, face components, and facial landmarks hierarchically. The regression process works in a unified cascaded fern framework with binary patterns. We present generalized gradient boosted ferns (GBFs) for the regression framework, which give better performance than ferns. The framework also achieves real time performance. We verify our method on the latest benchmark datasets and show that it achieves the state-of-the-art performance.