DTGen: Generative Diffusion-Based Few-Shot Data Augmentation for Fine-Grained Dirty Tableware Recognition

2025-09-15

AI Summary

To address the challenges of fine-grained classification and severe scarcity of real-world samples in cutlery soiling recognition—hindering industrial deployment—this paper proposes DTGen, the first generative diffusion model framework tailored for few-shot cutlery soiling recognition. Methodologically, it employs LoRA-efficient fine-tuning of diffusion models, leverages structured textual prompts to synthesize high-fidelity, diverse soiling images, and integrates CLIP-based cross-modal embeddings for automated quality filtering. Technically, DTGen pioneers the application of generative diffusion models to this industrial vision task, enabling effective fine-grained synthetic data augmentation; it supports edge-lightweight deployment and enables closed-loop optimization with dishwashing systems to reduce energy consumption and detergent usage. Experiments demonstrate substantial improvements in classification accuracy under extreme data scarcity (<10 real samples per class), validating the practical utility and engineering value of generative AI for intelligent food safety monitoring.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Intelligent tableware cleaning is a critical application in food safety and smart homes, but existing methods are limited by coarse-grained classification and scarcity of few-shot data, making it difficult to meet industrialization requirements. We propose DTGen, a few-shot data augmentation scheme based on generative diffusion models, specifically designed for fine-grained dirty tableware recognition. DTGen achieves efficient domain specialization through LoRA, generates diverse dirty images via structured prompts, and ensures data quality through CLIP-based cross-modal filtering. Under extremely limited real few-shot conditions, DTGen can synthesize virtually unlimited high-quality samples, significantly improving classifier performance and supporting fine-grained dirty tableware recognition. We further elaborate on lightweight deployment strategies, promising to transfer DTGen's benefits to embedded dishwashers and integrate with cleaning programs to intelligently regulate energy consumption and detergent usage. Research results demonstrate that DTGen not only validates the value of generative AI in few-shot industrial vision but also provides a feasible deployment path for automated tableware cleaning and food safety monitoring.

Problem

• Addresses fine-grained dirty tableware recognition with limited data
• Overcomes few-shot data scarcity using generative diffusion models
• Enables intelligent energy and detergent regulation in dishwashers

Innovation

• Generative diffusion models for data augmentation
• LoRA for efficient domain specialization
• CLIP-based cross-modal filtering quality