From Prompts to Deployment: Auto-Curated Domain-Specific Dataset Generation via Diffusion Models
This addresses the problem of data scarcity and distribution mismatch for practitioners deploying AI models in specific domains, though it is incremental as it builds on existing diffusion and validation techniques.
The paper tackles the distribution shift between pre-trained models and real-world deployment by developing an automated pipeline using diffusion models to generate domain-specific synthetic datasets, resulting in efficient construction of high-quality datasets that reduce reliance on extensive real-world data collection.
In this paper, we present an automated pipeline for generating domain-specific synthetic datasets with diffusion models, addressing the distribution shift between pre-trained models and real-world deployment environments. Our three-stage framework first synthesizes target objects within domain-specific backgrounds through controlled inpainting. The generated outputs are then validated via a multi-modal assessment that integrates object detection, aesthetic scoring, and vision-language alignment. Finally, a user-preference classifier is employed to capture subjective selection criteria. This pipeline enables the efficient construction of high-quality, deployable datasets while reducing reliance on extensive real-world data collection.