# Learning Facial Representations from the Cycle-consistency of Face

arXiv: https://arxiv.org/abs/2108.03427 · HTML: https://mlsanity.com/paper/2108.03427/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 52 references that are in the map.
- First posted on arXiv 7 Aug 2021.

## Builds on (most cited first, 22 of 22)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume](https://mlsanity.com/paper/1709.02371/) (2017)
- [VoxCeleb2: Deep Speaker Recognition](https://mlsanity.com/paper/1806.05622/) (2018)
- [SphereFace: Deep Hypersphere Embedding for Face Recognition](https://mlsanity.com/paper/1704.08063/) (2017)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)](https://mlsanity.com/paper/1703.07332/) (2017)
- [Challenges in Representation Learning: A report on three machine learning contests](https://mlsanity.com/paper/1307.0414/) (2013)
- [Few-Shot Adversarial Learning of Realistic Neural Talking Head Models](https://mlsanity.com/paper/1905.08233/) (2019)
- [X2Face: A network for controlling face generation by using images, audio, and pose codes](https://mlsanity.com/paper/1807.10550/) (2018)
- [GANimation: Anatomically-aware Facial Animation from a Single Image](https://mlsanity.com/paper/1807.09251/) (2018)
- [Face Alignment in Full Pose Range: A 3D Total Solution](https://mlsanity.com/paper/1804.01005/) (2018)
- [Unsupervised Discovery of Object Landmarks as Structural Representations](https://mlsanity.com/paper/1804.04412/) (2018)
- [Neural Face Editing with Intrinsic Image Disentangling](https://mlsanity.com/paper/1704.04131/) (2017)
- [Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance](https://mlsanity.com/paper/1806.06503/) (2018)
- [Disentangled Representation Learning for 3D Face Shape](https://mlsanity.com/paper/1902.09887/) (2019)
- [Self-supervised learning of a facial attribute embedding from video](https://mlsanity.com/paper/1808.06882/) (2018)
- [Learning Residual Images for Face Attribute Manipulation](https://mlsanity.com/paper/1612.05363/) (2016)
- [MirrorFlow: Exploiting Symmetries in Joint Optical Flow and Occlusion Estimation](https://mlsanity.com/paper/1708.05355/) (2017)
- [Unsupervised Learning of Face Representations](https://mlsanity.com/paper/1803.01260/) (2018)

## Built on by (most cited first, 20 of 20)

- [SVFAP: Self-supervised Video Facial Affect Perceiver](https://mlsanity.com/paper/2401.00416/) (2023)
- [Pose-disentangled Contrastive Learning for Self-supervised Facial Representation](https://mlsanity.com/paper/2211.13490/) (2022)
- [Self-Supervised Facial Representation Learning with Facial Region Awareness](https://mlsanity.com/paper/2403.02138/) (2024)
- [Revisiting Self-Supervised Contrastive Learning for Facial Expression Recognition](https://mlsanity.com/paper/2210.03853/) (2022)
- [Exploring Facial Expression Recognition through Semi-Supervised Pretraining and Temporal Modeling](https://mlsanity.com/paper/2403.11942/) (2024)
- [Diff-ID: An Explainable Identity Difference Quantification Framework for DeepFake Detection](https://mlsanity.com/paper/2303.18174/) (2023)
- [Exploring Large-scale Unlabeled Faces to Enhance Facial Expression Recognition](https://mlsanity.com/paper/2303.08617/) (2023)
- [Unsupervised Facial Expression Representation Learning with Contrastive Local Warping](https://mlsanity.com/paper/2303.09034/) (2023)
- [Emotic Masked Autoencoder with Attention Fusion for Facial Expression Recognition](https://mlsanity.com/paper/2403.13039/) (2024)
- [Unpaired Multi-domain Attribute Translation of 3D Facial Shapes with a Square and Symmetric Geometric Map](https://mlsanity.com/paper/2308.13245/) (2023)
- [On Biased Behavior of GANs for Face Verification](https://mlsanity.com/paper/2208.13061/) (2022)
- [A Survey of Face Recognition](https://mlsanity.com/paper/2212.13038/) (2022)
- [A Generative Framework for Self-Supervised Facial Representation Learning](https://mlsanity.com/paper/2309.08273/) (2023)
- [LH2Face: Loss function for Hard High-quality Face](https://mlsanity.com/paper/2506.23555/) (2025)
- [Towards an objective characterization of an individual's facial movements using Self-Supervised Person-Specific-Models](https://mlsanity.com/paper/2211.08279/) (2022)
- [Watch Out for the Confusing Faces: Detecting Face Swapping with the Probability Distribution of Face Identification Models](https://mlsanity.com/paper/2303.13131/) (2023)
- [DrFER: Learning Disentangled Representations for 3D Facial Expression Recognition](https://mlsanity.com/paper/2403.08318/) (2024)
- [eMotion-GAN: A Motion-based GAN for Photorealistic and Facial Expression Preserving Frontal View Synthesis](https://mlsanity.com/paper/2404.09940/) (2024)
- [Learn Your Scales: Towards Scale-Consistent Generative Novel View Synthesis](https://mlsanity.com/paper/2503.15412/) (2025)
- [Semi-Supervised Facial Expression Recognition based on Dynamic Threshold and Negative Learning](https://mlsanity.com/paper/2601.05556/) (2026)

## Abstract (arXiv)

Faces manifest large variations in many aspects, such as identity, expression, pose, and face styling. Therefore, it is a great challenge to disentangle and extract these characteristics from facial images, especially in an unsupervised manner. In this work, we introduce cycle-consistency in facial characteristics as free supervisory signal to learn facial representations from unlabeled facial images. The learning is realized by superimposing the facial motion cycle-consistency and identity cycle-consistency constraints. The main idea of the facial motion cycle-consistency is that, given a face with expression, we can perform de-expression to a neutral face via the removal of facial motion and further perform re-expression to reconstruct back to the original face. The main idea of the identity cycle-consistency is to exploit both de-identity into mean face by depriving the given neutral face of its identity via feature re-normalization and re-identity into neutral face by adding the personal attributes to the mean face. At training time, our model learns to disentangle two distinct facial representations to be useful for performing cycle-consistent face reconstruction. At test time, we use the linear protocol scheme for evaluating facial representations on various tasks, including facial expression recognition and head pose regression. We also can directly apply the learnt facial representations to person recognition, frontalization and image-to-image translation. Our experiments show that the results of our approach is competitive with those of existing methods, demonstrating the rich and unique information embedded in the disentangled representations. Code is available at https://github.com/JiaRenChang/FaceCycle .