Get Started
Home
Topics
Search
Library
Research questionHow can recognition models be augmented with synthetic data when external foundation models and datasets are unavailable?Recognition systems may need varied or difficult training examples, but policy, legal, or access restrictions can rule out external foundation models and datasets. The available data must therefore support augmentation without assuming outside resources.
AI
Computer Vision
Diffusion Models
Machine Learning
Research Paper
Latest papersRecent research connected to this question, newest first.ScoreMix: Synthetic Data Generation by Score Composition in Diffusion Models Improves RecognitionThe source studies a self-contained diffusion-based generator that mixes class-conditioned scores during reverse diffusion to create hard samples for recognition. It examines class selection using discriminator embedding distances and reports results on eight public face-recognition benchmarks; its evidence is limited to those experiments.research paper · Sep 3, 2026
Related questions
How can synthetic image augmentation reliably improve computer vision when labeled real data are scarce?How can we generate realistic synthetic identity documents for verification evaluation when real documents are scarce?When augmentation quantity is fixed, how does synthetic-example placement in representation space affect imbalanced pragmatic-function classification?How can MEG foundation models become reusable across subjects, sites, and tasks despite limited training data?