Learning Unified Representations for Multi-Resolution Face Recognition

Hulingxiao He, Wu Yuan, Yidian Huang, Shilong Zhao, Wen Yuan, HanQing Li

Keywords: multi-resolution face recognition, deep representation learning

TL;DR: We propose Branch-to-Trunk Network to learn discriminative embeddings for multi-resolution face recognition while preserving representation compatibility.

Abstract: In this work, we propose Branch-to-Trunk network (BTNet), a novel representation learning method for multi-resolution face recognition. It consists of a trunk network (TNet), namely a unified encoder, and multiple branch networks (BNets), namely resolution adapters. As per the input, a resolution-specific BNet is used and the output are implanted as feature maps in the feature pyramid of TNet, at a layer with the same resolution. The discriminability of tiny faces is significantly improved, as the interpolation error introduced by rescaling, especially up-sampling, is mitigated on the inputs. With branch distillation and backward-compatible training, BTNet transfers discriminative high-resolution information to multiple branches while guaranteeing representation compatibility. Our experiments demonstrate strong performance on face recognition benchmarks, both for multi-resolution identity matching and feature aggregation, with much less computation amount and parameter storage. We establish new state-of-the-art on the challenging QMUL-SurvFace 1: N face identification task.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/learning-unified-representations-for-multi/code)