A Novel Multi-Task Tensor Correlation Neural Network for Facial Attribute Prediction
Preprint
Abstract
Face multi-attribute prediction benefits substantially from multi-task learning (MTL), which learns multiple face attributes simultaneously to achieve shared or mutually related representations of different attributes. The most widely used MTL convolutional neural network is heuristically or empirically designed by sharing all of the convolutional layers and splitting at the fully connected layers for task-specific losses. However, it is improper to view all low and mid-level features for different attributes as being the same, especially when these attributes are only loosely related. In this paper, we propose a novel multi-attribute tensor correlation neural network (MTCN) for face attribute prediction. The structure shares the information in low-level features (e.g., the first two convolutional layers) but splits that in high-level features (e.g., from the third convolutional layer to the fully connected layer). At the same time, during high-level feature extraction, each subnetwork (e.g., Age-Net, Gender-Net, ..., and Smile-Net) excavates closely related features from other networks to enhance its features. Then, we project the features of the C9 layers of the fine-tuned subnetworks into a highly correlated space by using a novel tensor correlation analysis algorithm (NTCCA). The final face attribute prediction is made based on the correlation matrix. Experimental results on benchmarks with multiple face attributes (CelebA and LFWA) show that the proposed approach has superior performance compared to state-of-the-art methods.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- AFFACT: Alignment-free facial attribute classification technique2017
- Deep Expectation of Real and Apparent Age from a Single Image Without Facial LandmarksInternational Journal of Computer Vision, 2016
- Multi-Task CNN Model for Attribute PredictionIEEE Transactions on Multimedia, 2015
- Age Estimation via Grouping and Decision FusionIEEE Transactions on Information Forensics and Security, 2015
- Tensor Canonical Correlation Analysis for Multi-View Dimension ReductionIEEE Transactions on Knowledge and Data Engineering, 2015
- Combining Facial Dynamics With Appearance for Age EstimationIEEE Transactions on Image Processing, 2015
- Exploring Context and Content Links in Social Media: A Latent Space MethodIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Scalable Face Image Retrieval with Identity-Based Quantization and Multireference RerankingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Sharing features between objects and their attributes2011
- Age Synthesis and Estimation via Faces: A SurveyIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010