The field of face recognition and generation is rapidly advancing, with a focus on improving accuracy, efficiency, and privacy. Recent developments have seen the introduction of new methods for face swapping, face aging, and face de-aging, as well as advancements in identity-preserving video generation and diffusion-based face generation. Notably, researchers have proposed novel approaches to address challenges such as out-of-gallery detection, face quality assessment, and privacy-preserving face recognition. These innovations have the potential to significantly impact various applications, including security, entertainment, and healthcare. Noteworthy papers include: LaVieID, which presents a local autoregressive diffusion transformer for identity-preserving video creation, and NegFaceDiff, which introduces a novel sampling method that incorporates negative conditions into the identity-conditioned diffusion process to enhance identity separation.
Advances in Face Recognition and Generation
Sources
Are you In or Out (of gallery)? Wisdom from the Same-Identity Crowd
MotionSwap
Parity Requires Unified Input Dependence and Negative Eigenvalues in SSMs
LaVieID: Local Autoregressive Diffusion Transformers for Identity-Preserving Video Creation
MSPT: A Lightweight Face Image Quality Assessment Method with Multi-stage Progressive Training
ShoulderShot: Generating Over-the-Shoulder Dialogue Videos
DiTVR: Zero-Shot Diffusion Transformer for Video Restoration
Stand-In: A Lightweight and Plug-and-Play Identity Control for Video Generation
VOIDFace: A Privacy-Preserving Multi-Network Face Recognition With Enhanced Security
StableAvatar: Infinite-Length Audio-Driven Avatar Video Generation
SelfHVD: Self-Supervised Handheld Video Deblurring for Mobile Phones
Identity-Preserving Aging and De-Aging of Faces in the StyleGAN Latent Space
Synthetic Data Generation for Emotional Depth Faces: Optimizing Conditional DCGANs via Genetic Algorithms in the Latent Space and Stabilizing Training with Knowledge Distillation
Personalized Feature Translation for Expression Recognition: An Efficient Source-Free Domain Adaptation Method
Gen-AFFECT: Generation of Avatar Fine-grained Facial Expressions with Consistent identiTy
From Large Angles to Consistent Faces: Identity-Preserving Video Generation via Mixture of Facial Experts
NegFaceDiff: The Power of Negative Context in Identity-Conditioned Diffusion for Synthetic Face Generation
Enhancing Diffusion Face Generation with Contrastive Embeddings and SegFormer Guidance
Trajectory-aware Shifted State Space Models for Online Video Super-Resolution
Hybrid Generative Fusion for Efficient and Privacy-Preserving Face Recognition Dataset Generation