In real world applications, multi-view data has attracted intensive attention due to the complex and complementary relationship across views. Multi-view representation learning (MvRL) focuses on ...

To address this issue, we propose adaptive graph representation learning and reasoning over facial components, aiming to learn representative vertices that describe each component, exploit the ...