README.md

General Information
Repository Size and Activity
Contribution Statistics
Other Metrics
Application
Progress Status
Main

---

FG 2024 Papers: Explore a comprehensive collection of innovative research papers presented at FG 2024, one of the premier conferences on automatic face and gesture recognition. Seamlessly integrate code implementations for better understanding. ⭐ Experience the cutting edge of progress in facial analysis, gesture recognition, and biometrics with this repository!

:point_right: * This count includes repositories on GitHub, GitLab, Hugging Face, and distributions on PyPI, while excluding Web Page or GitHub Page links.

---

---

Other collections of the best AI conferences

Conference	Year
	2023	2024
Computer Vision (CV)
CVPR
ICCV
ECCV
WACV	:heavy_minus_sign:
FG	:heavy_minus_sign:
Speech/Signal Processing (SP/SigProc)
ICASSP
INTERSPEECH
ISMIR		:heavy_minus_sign:
Natural Language Processing (NLP)
EMNLP
Machine Learning (ML)
AAAI	:heavy_minus_sign:
ICLR	:heavy_minus_sign:
ICML	:heavy_minus_sign:
NeurIPS	:heavy_minus_sign:

---

Contributors

---

Papers

List of sections

• Best Reviewed Papers
• Best Reviewed Student Papers
• Face Biometrics
• Facial Expressions
• Human Pose and Motion
• Gait and Action
• Hand and Sign Language
• Animation, Synthesis and Self-Supervision
• Posters
• Demo presentation

Best Reviewed Papers

Title
An Active-Gaze Morphable Model for 3D Gaze Estimation	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Occluded Person Retrieval with Hierarchical Feature Optimization	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
High-Resolution Image Enumeration for Low-Resolution Face Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
OpenThermalPose: An Open-Source Annotated Thermal Human Pose Dataset and Initial YOLOv8-Pose Baselines	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:

Best Reviewed Student Papers

Title
A Unified Model for Gaze Following and Social Gaze Prediction	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
DrFER: Learning Disentangled Representations for 3D Facial Expression Recognition	:heavy_minus_sign:		:heavy_minus_sign:
ClipSwap: Towards High Fidelity Face Swapping via Attribute and CLIP-Informed Loss	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Multi-Modal Human Behaviour Graph Representation Learning for Automatic Depression Assessment	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:

Face Biometrics

Title
Designing Cross-Race Tests for Forensic Facial Examiners, Super-Recognizers, and Face Recognition Algorithm	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
TetraLoss: Improving the Robustness of Face Recognition against Morphing Attacks	:heavy_minus_sign:		:heavy_minus_sign:
Hierarchical Generative Network for Face Morphing Attacks	:heavy_minus_sign:		:heavy_minus_sign:
Face Anti-Spoofing via Interaction Learning with Face Image Quality Alignment	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:

Facial Expressions

Title
Multi-Scale Spatio-Temporal Graph Convolutional Network for Facial Expression Spotting	:heavy_minus_sign:		:heavy_minus_sign:
epsilon-Mesh Attack: A Surface-based Adversarial Point Cloud Attack for Facial Expression Recognition			:heavy_minus_sign:
Distilling Privileged Multimodal Information for Expression Recognition using Optimal Transport			:heavy_minus_sign:
CSTalk: Correlation Supervised Speech-Driven 3D Emotional Facial Animation Generation	:heavy_minus_sign:		:heavy_minus_sign:

Human Pose and Motion

Title
Uncalibrated Multi-View 3D Human Pose Estimation with Geometry Driven Attention	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Geometry-Biased Transformer for Robust Multi-View 3D Human Pose Reconstruction	:heavy_minus_sign:		:heavy_minus_sign:
BEAVP: A Bidirectional Enhanced Adversarial Model for Video Prediction	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
CasCalib: Cascaded Calibration for Motion Capture from Sparse Unsynchronized Cameras			:heavy_minus_sign:

Gait and Action

Title
Unveiling Gender Effects in Gait Recognition using Conditional-Matched Bootstrap Analysis		:heavy_minus_sign:	:heavy_minus_sign:
GaitPT: Skeletons are All You Need for Gait Recognition	:heavy_minus_sign:		:heavy_minus_sign:
Attention Prompt Tuning: Parameter-Efficient Adaptation of Pre-trained Models for Action Recognition			:heavy_minus_sign:
ViewDiffGait: View Pyramid Diffusion for Gait Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:

Hand and Sign Language

Title
Two Hands are Better than One: Resolving Hand to Hand Intersections via Occupancy Networks	:heavy_minus_sign:		:heavy_minus_sign:
SynthSL: Expressive Humans for Sign Language Image Synthesis	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
A Gloss-Free Sign Language Production with Discrete Representation	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
In My Perspective, in My Hands: Accurate Egocentric 2D Hand Pose and Action Recognition			:heavy_minus_sign:

Animation, Synthesis and Self-Supervision

Title
EAT-Face: Emotion-Controllable Audio-Driven Talking Face Generation via Diffusion Model	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Embedded Representation Learning Network for Animating Styled Video Portrait	:heavy_minus_sign:		:heavy_minus_sign:
Giving a Hand to Diffusion Models: A Two-Stage Approach to Improving Conditional Human Image Generation			:heavy_minus_sign:
RS-rPPG: Robust Self-Supervised Learning for rPPG		:heavy_minus_sign:	:heavy_minus_sign:

Posters

Title
Efficient Verification-based Face Identification	:heavy_minus_sign:		:heavy_minus_sign:
Dataset Infant Anonymization with Pose and Emotion Retention	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Face the Needle: Predicting Risk of Fear and Fainting During Blood Donation through Video Analysis	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Intra-Person Camera Adversarial for Intra-Camera Supervised Person Re-Identification	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Adaptive Cross-Architecture Mutual Knowledge Distillation	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
ASPECD: Adaptable Soft-Biometric Privacy-Enhancement using Centroid Decoding for Face Verification	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Young Labeled Faces in the Wild (YLFW): A Dataset for Children Faces Recognition			:heavy_minus_sign:
Deepfake: Classifiers, Fairness, and Demographically Robust Algorithm	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
PointFaceFormer: Local and Global Attention based Transformer for 3D Point Cloud Face Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Subject-based Domain Adaptation for Facial Expression Recognition			:heavy_minus_sign:
Efficient Detection of Disguised Faces using Photos/Sketches from Low-Quality Surveillance Footage	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Lip and Speech Synchronization using Supervised Contrastive Learning and Cross-Modal Attention	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
If it's not Enough, Make it so: Reducing Authentic Data Demand in Face Recognition through Synthetic Faces	:heavy_minus_sign:		:heavy_minus_sign:
Data Augmentation Techniques for Enhanced Facial Landmarks Detection in Patients with Repaired Cleft Lip and Palate	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Deep Adaptative Spectral Zoom for Improved Remote Heart Rate Estimation	:heavy_minus_sign:		:heavy_minus_sign:
Bridging the Gap: Protocol Towards Fair and Consistent Affect Analysis			:heavy_minus_sign:
ONOT: a High-Quality ICAO-Compliant Synthetic Mugshot Dataset			:heavy_minus_sign:
RFIS-FPI: Reversible Face Image Steganography Neural Network for Face Privacy Interactions	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Unlocking the Black Box: Concept-based Modeling for Interpretable Affective Computing Applications	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Social-MAE: A Transformer-based Multimodal Autoencoder for Face and Voice	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Guided Interpretable Facial Expression Recognition via Spatial Action Unit Cues			:heavy_minus_sign:
AerialFace: A Light Weight Framework for Unmanned Aerial Vehicle Face Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
QGFace: Quality-Guided Joint Training for Mixed Quality Face Recognition			:heavy_minus_sign:
EmoCLIP: A Vision-Language Method for Zero-Shot Video Facial Expression Recognition			:heavy_minus_sign:
In-Domain Inversion for Improved 3D Face Alignment on Asymmetrical Expressions	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
3D Face Modeling via Weakly-Supervised Disentanglement Network Joint Identity-Consistency Prior			:heavy_minus_sign:
Expression-Aware Masking and Progressive Decoupling for Cross-Database Facial Expression Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Explainable Face Verification via Feature-Guided Gradient Backpropagation	:heavy_minus_sign:		:heavy_minus_sign:
One-Stage Open-Vocabulary Temporal Action Detection Leveraging Temporal Multi-Scale and Action Label Features	:heavy_minus_sign:		:heavy_minus_sign:
Skeleton-based Self-Supervised Feature Extraction for Improved Dynamic Hand Gesture Recognition		:heavy_minus_sign:	:heavy_minus_sign:
Human Action Recognition with Multi-Level Granularity and Pair-Wise Hyper GCN	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
MGRFormer: A Multimodal Transformer Approach for Surgical Gesture Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
CCDb-HG: Novel Annotations and Gaze-Aware Representations for Head Gesture Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
GestSpoof: Gesture based Spatio-Temporal Representation Learning for Robust Fingerprint Presentation Attack Detection	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Spatio Temporal Sparse Graph Convolution Network for Hand Gesture Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Crowd Detection via Point Localization with Diffusion Models	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
MIMIC-Pose: Implicit Membership Discrimination of Body Joints for Human Pose Estimation	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
DPA-2D: Depth Propagation and Alignment with 2D Observations Guidance for Human Mesh Recovery	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Evaluating Recent 2D Human Pose Estimators for 2D-3D Pose Lifting		:heavy_minus_sign:	:heavy_minus_sign:
The Paradox of Motion: Evidence for Spurious Correlations in Skeleton-based Gait Recognition Models	:heavy_minus_sign:		:heavy_minus_sign:
Improving 2D Human Pose Estimation in Unseen Camera Views with Synthetic Data	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
DualH: A Dual Hierarchical Model for Temporal Action Localization		:heavy_minus_sign:	:heavy_minus_sign:
HR-xNet: A Novel High-Resolution Network for Human Pose Estimation with Low Resource Consumption	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Cross-Block Fine-Grained Semantic Cascade for Skeleton-based Sports Action Recognition	:heavy_minus_sign:		:heavy_minus_sign:
HM-Auth: Redefining User Authentication in Immersive Virtual World through Hand Movement Signatures	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
A Data-Driven Representation for Sign Language Production			:heavy_minus_sign:
Diversity-Aware Sign Language Production through a Pose Encoding Variational Autoencoder	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Resource-Efficient Gesture Recognition using Low-Resolution Thermal Camera via Spiking Neural Networks and Sparse Segmentation	:heavy_minus_sign:		:heavy_minus_sign:
Transfer Learning for Cross-Dataset Isolated Sign Language Recognition in Under-Resourced Datasets			:heavy_minus_sign:
Patch-based Privacy Attention for Weakly-Supervised Privacy-Preserving Action Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Boosting Gesture Recognition with an Automatic Gesture Annotation Framework	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Towards Better Communication: Refining Hand Pose Estimation in Low-Resolution Sign Language Videos	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Quantifying Biometric Characteristics of Hand Gestures through Feature Space Probing and Identity-Level Cross-Gesture Disentanglement	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Hand Graph Topology Selection for Skeleton-based Sign Language Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Unconstrained Hand Recognition using Thermal Infrared Sensing of Dorsal Veins	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Integrating a Hierarchical Structure of Situated Human Motion in Multi-Task Learning for Professional Gesture Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Towards High Fidelity and Accurate Face Swapping	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Face-based Strategies for Evaluating Asymmetry and Speech Articulation in Patients with Craniofacial Anomalies	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Audio-Visual Person Verification based on Recursive Fusion of Joint Cross-Attention			:heavy_minus_sign:
VoxAtnNet: A 3D Point Clouds Convolutional Neural Network for Generalizable Face Presentation Attack Detection	:heavy_minus_sign:		:heavy_minus_sign:
Multi-View Consistent 3D GAN Inversion via Bidirectional Encoder		:heavy_minus_sign:	:heavy_minus_sign:
Context-based Dataset for Analysis of Videos of Autistic Children	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Seeing and Hearing what has not been Said; A Multimodal Client Behavior Classifier in Motivational Interviewing with Interpretable Fusion	:heavy_minus_sign:		:heavy_minus_sign:
SignAvatar: Sign Language 3D Motion Reconstruction and Generation			:heavy_minus_sign:
PortraitDAE: Line-Drawing Portraits Style Transfer from Photos via Diffusion Autoencoder with Meaningful Encoded Noise	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
FE-Adapter: Adapting Image-based Emotion Classifiers to Videos	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Latent Embedding Clustering for Occlusion Robust Head Pose Estimation	:heavy_minus_sign:		:heavy_minus_sign:
Pivotal Tuning Editing: Towards Disentangled Wrinkle Editing with GANs	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Data-Driven but Privacy-Conscious: Pedestrian Dataset De-Identification via Full-Body Person Synthesis	:heavy_minus_sign:		:heavy_minus_sign:
CrossGaze: A Strong Method for 3D Gaze Estimation in the Wild	:heavy_minus_sign:		:heavy_minus_sign:
Survey of Automated Methods for Nonverbal Behavior Analysis in Parent-Child Interactions	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Naive Data Augmentation Might be Toxic: Data-Prior Guided Self-Supervised Representation Learning for Micro-Gesture Recognition	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
SMCTL: Subcarrier Masking Contrastive Transfer Learning for Human Gesture Recognition with Passive Wi-Fi Sensing	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Semantic-Aware Detail Enhancement for Blind Face Restoration	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Discovering Interpretable Directions in the Semantic Latent Space of Diffusion Models			:heavy_minus_sign:
Breaking Template Protection: Reconstruction of Face Images from Protected Facial Templates	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Benchmarking Skeleton-based Motion Encoder Models for Clinical Applications: Estimating Parkinson's Disease Severity in Walking Sequences	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
The Seven Faces of Stress: Understanding Facial Activity Patterns during Cognitive Stress	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Visual Coherence Face Anonymization Algorithm based on Dynamic Identity Perception	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
PyraMoT: A Novel Framework for Enhanced Facial Thermal Landmarks Detection	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Visual Saliency Guided Gaze Target Estimation with Limited Labels	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Hyp-OC: Hyperbolic One Class Classifier for Face Anti-Spoofing	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Dynamic Cross Attention for Audio-Visual Person Verification			:heavy_minus_sign:
Enhancing Privacy in Face Analytics using Fully Homomorphic Encryption	:heavy_minus_sign:		:heavy_minus_sign:
CribNet: Enhancing Infant Safety in Cribs through Vision-based Hazard Detection		:heavy_minus_sign:	:heavy_minus_sign:
3D Face Morphing Attack Generation using Non-Rigid Registration	:heavy_minus_sign:		:heavy_minus_sign:
BTVSL: A Novel Sentence-Level Annotated Dataset for Bangla Sign Language Translation	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:

Demo presentation

Title
Russian Sign Language Learning Simulator	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:
Expanding PyAFAR: A Novel Privacy-Preserving Infant AU Detector	:heavy_minus_sign:	:heavy_minus_sign:	:heavy_minus_sign:

---

Key Terms

---

Star History