Multicolumn Networks for Face Recognition

Abstract

The objective of this work is set-based face recognition, i.e. to decide if two sets of images of a face are of the same person or not. Conventionally, the set-wise feature descriptor is computed as an average of the descriptors from individual face images within the set. In this paper, we design a neural network architecture that learns to aggregate based on both "visual" quality (resolution, illumination), and "content" quality (relative importance for discriminative classification). To this end, we propose a Multicolumn Network (MN) that takes a set of images (the number in the set can vary) as input, and learns to compute a fix-sized feature descriptor for the entire set. To encourage high-quality representations, each individual input image is first weighted by its "visual" quality, determined by a self-quality assessment module, and followed by a dynamic recalibration based on "content" qualities relative to the other images within the set. Both of these qualities are learnt implicitly during training for set-wise classification. Comparing with the previous state-of-the-art architectures trained with the same dataset (VGGFace2), our Multicolumn Networks show an improvement of between 2-6% on the IARPA IJB face recognition benchmarks, and exceed the state of the art for all methods on these benchmarks.

To appear in BMVC2018

References in corpus (2)

• Learning Face Representation from Scratch
• A simple neural network module for relational reasoning

Cited by in corpus (23)

• SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition
• Finger-GAN: Generating Realistic Fingerprint Images Using Connectivity Imposed GAN
• CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognition
• Iris-GAN: Learning to Generate Realistic Iris Images Using Convolutional GAN
• Face Image Quality Assessment: A Literature Survey
• SphereFace2: Binary Classification is All You Need for Deep Face Recognition
• PoseFace: Pose-Invariant Features and Pose-Adaptive Loss for Face Recognition
• Jointly De-biasing Face Recognition and Demographic Attribute Estimation
• Recurrent Embedding Aggregation Network for Video Face Recognition
• Inducing Predictive Uncertainty Estimation for Face Recognition
• Towards Universal Representation Learning for Deep Face Recognition
• Data Uncertainty Learning in Face Recognition
• Attention Control with Metric Learning Alignment for Image Set-based Recognition
• Towards Interpretable Face Recognition
• Fast and Reliable Probabilistic Face Embeddings in the Wild
• Multi-Prototype Networks for Unconstrained Set-based Face Recognition
• GhostVLAD for set-based face recognition
• MultiFace: A Generic Training Mechanism for Boosting Face Recognition Performance
• Discriminability Distillation in Group Representation Learning
• Improving Face Recognition from Hard Samples via Distribution Distillation Loss
• Set-Based Face Recognition Beyond Disentanglement: Burstiness Suppression With Variance Vocabulary
• 3D-Aided Data Augmentation for Robust Face Understanding
• On Improving the Generalization of Face Recognition in the Presence of Occlusions