# Hybrid Generative Fusion for Efficient and Privacy-Preserving Face Recognition Dataset Generation arXiv: https://arxiv.org/abs/2508.10672 ยท HTML: https://mlsanity.com/paper/2508.10672/ ## Key facts - Cited by 5 papers in the map (data through 14 Oct 2026). - Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. - Builds on 30 of its 38 references that are in the map. - First posted on arXiv 14 Aug 2025. ## Placed in - Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/) - Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/) ## Builds on (most cited first, 30 of 30) - [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021) - [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023) - [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023) - [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020) - [Towards Fast, Accurate and Stable 3D Dense Face Alignment](https://mlsanity.com/paper/2009.09960/) (2020) - [DCFace: Synthetic Face Generation with Dual Condition Diffusion Model](https://mlsanity.com/paper/2304.07060/) (2023) - [SynFace: Face Recognition with Synthetic Data](https://mlsanity.com/paper/2108.07960/) (2021) - [Arc2Face: A Foundation Model for ID-Consistent Human Faces](https://mlsanity.com/paper/2403.11641/) (2024) - [IDiff-Face: Synthetic-based Face Recognition through Fizzy Identity-Conditioned Diffusion Models](https://mlsanity.com/paper/2308.04995/) (2023) - [SFace: Privacy-friendly and Accurate Face Recognition using Synthetic Data](https://mlsanity.com/paper/2206.10520/) (2022) - [ReconBoost: Boosting Can Achieve Modality Reconcilement](https://mlsanity.com/paper/2405.09321/) (2024) - [SDFR: Synthetic Data for Face Recognition Competition](https://mlsanity.com/paper/2404.04580/) (2024) - [Second Edition FRCSyn Challenge at CVPR 2024: Face Recognition Challenge in the Era of Synthetic Data](https://mlsanity.com/paper/2404.10378/) (2024) - [AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation](https://mlsanity.com/paper/2409.20398/) (2024) - [Image Fusion for Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2502.15694/) (2024) - [Prompt Categories Cluster for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2412.13823/) (2024) - [One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Framework](https://mlsanity.com/paper/2505.11131/) (2025) - [Logical Consistency and Greater Descriptive Power for Facial Hair Attribute Learning](https://mlsanity.com/paper/2302.11102/) (2023) - [Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors](https://mlsanity.com/paper/2409.02979/) (2024) - [Image Augmentation Agent for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2412.20439/) (2024) - [What Should Be Balanced in a "Balanced" Face Recognition Dataset?](https://mlsanity.com/paper/2304.09818/) (2023) - [OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning](https://mlsanity.com/paper/2505.05180/) (2025) - [MOL: Joint Estimation of Micro-Expression, Optical Flow, and Landmark via Transformer-Graph-Style Convolution](https://mlsanity.com/paper/2506.14511/) (2025) - [Consistency and Accuracy of CelebA Attribute Values](https://mlsanity.com/paper/2210.07356/) (2022) - [Top-K Pooling with Patch Contrastive Learning for Weakly-Supervised Semantic Segmentation](https://mlsanity.com/paper/2310.09828/) (2023) - [Not All Pairs are Equal: Hierarchical Learning for Average-Precision-Oriented Video Retrieval](https://mlsanity.com/paper/2407.15566/) (2024) - [Improved Diversity-Promoting Collaborative Metric Learning for Recommendation](https://mlsanity.com/paper/2409.01012/) (2024) - [Facial Action Unit Detection via Adaptive Attention and Relation](https://mlsanity.com/paper/2001.01168/) (2020) - [Facial Action Unit Detection by Adaptively Constraining Self-Attention and Causally Deconfounding Sample](https://mlsanity.com/paper/2410.01251/) (2024) - [MixBridge: Heterogeneous Image-to-Image Backdoor Attack through Mixture of Schr"odinger Bridges](https://mlsanity.com/paper/2505.08809/) (2025) ## Built on by (most cited first, 5 of 5) - [RoboRouter: Training-Free Policy Routing for Robotic Manipulation](https://mlsanity.com/paper/2603.07892/) (2026) - [Dual-Stage Reweighted MoE for Long-Tailed Egocentric Mistake Detection](https://mlsanity.com/paper/2509.12990/) (2025) - [Guiding Diffusion-based Reconstruction with Contrastive Signals for Balanced Visual Representation](https://mlsanity.com/paper/2603.04803/) (2026) - [Divide, Weight, and Route: Difficulty-Aware Optimization with Dynamic Expert Fusion for Long-tailed Recognition](https://mlsanity.com/paper/2508.19630/) (2025) - [Deconstructing the Failure of Ideal Noise Correction: A Three-Pillar Diagnosis](https://mlsanity.com/paper/2603.12997/) (2026) ## Abstract (arXiv) In this paper, we present our approach to the DataCV ICCV Challenge, which centers on building a high-quality face dataset to train a face recognition model. The constructed dataset must not contain identities overlapping with any existing public face datasets. To handle this challenge, we begin with a thorough cleaning of the baseline HSFace dataset, identifying and removing mislabeled or inconsistent identities through a Mixture-of-Experts (MoE) strategy combining face embedding clustering and GPT-4o-assisted verification. We retain the largest consistent identity cluster and apply data augmentation up to a fixed number of images per identity. To further diversify the dataset, we generate synthetic identities using Stable Diffusion with prompt engineering. As diffusion models are computationally intensive, we generate only one reference image per identity and efficiently expand it using Vec2Face, which rapidly produces 49 identity-consistent variants. This hybrid approach fuses GAN-based and diffusion-based samples, enabling efficient construction of a diverse and high-quality dataset. To address the high visual similarity among synthetic identities, we adopt a curriculum learning strategy by placing them early in the training schedule, allowing the model to progress from easier to harder samples. Our final dataset contains 50 images per identity, and all newly generated identities are checked with mainstream face datasets to ensure no identity leakage. Our method achieves 1st place in the competition, and experimental results show that our dataset improves model performance across 10K, 20K, and 100K identity scales. Code is available at https://github.com/Ferry-Li/datacvfr.