Research on face recognition has had a revival in recent years. This is largely due to the need for robust face recognition technologies for consumers to tag digital photos and facilitate their organization and online sharing. Face recognition in consumer digital imaging suffers from multiple factors such as uncontrolled lighting, large pose variation, a range of facial expressions, make-up, and severe partial occlusions, etc. These factors can be modeled by constructing a multi-linear structure based on a high-order tensor. In the other hand, many researches proved the high dimensional facial image lies on a low dimensional manifold structure. I will devote myself to find the low dimensional manifold structure of the facial image tensor representations. Recently, spare representation is a hot topic in pattern recognition. It has a good performance on image alignment, dimensional reduction, etc. I will use spare representation on my further research. I hope that it can obtain better performance on the uncontrolled face recognition. Meanwhile, I also focus on age estimation, action recognition, iris recognition and multimodal biometric identification. I enjoy the above research.