SynSpill: Improved Industrial Spill Detection With Synthetic Data

2025-08-13

AI Summary

In industrial leak detection, the scarcity of real annotated data—exacerbated by the rarity and sensitivity of leakage events—severely degrades zero-shot performance of vision-language models (VLMs), while conventional fine-tuning remains infeasible due to data constraints. To address this, we propose SynSpill, a scalable, high-fidelity synthetic data generation framework tailored for safety-critical scenarios, coupled with parameter-efficient fine-tuning (PEFT) strategies adaptable to diverse architectures—including YOLO, DETR, and large-scale VLMs. SynSpill effectively bridges domain gaps, significantly enhancing both zero-shot and few-shot generalization without requiring real annotations. Extensive experiments demonstrate consistent and comparable performance gains across all evaluated detectors and VLMs, validating the method’s effectiveness, robustness, and cross-architectural scalability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Large Vision Models
• Natural Language Processing: Safety and Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large-scale Vision-Language Models (VLMs) have transformed general-purpose visual recognition through strong zero-shot capabilities. However, their performance degrades significantly in niche, safety-critical domains such as industrial spill detection, where hazardous events are rare, sensitive, and difficult to annotate. This scarcity -- driven by privacy concerns, data sensitivity, and the infrequency of real incidents -- renders conventional fine-tuning of detectors infeasible for most industrial settings. We address this challenge by introducing a scalable framework centered on a high-quality synthetic data generation pipeline. We demonstrate that this synthetic corpus enables effective Parameter-Efficient Fine-Tuning (PEFT) of VLMs and substantially boosts the performance of state-of-the-art object detectors such as YOLO and DETR. Notably, in the absence of synthetic data (SynSpill dataset), VLMs still generalize better to unseen spill scenarios than these detectors. When SynSpill is used, both VLMs and detectors achieve marked improvements, with their performance becoming comparable. Our results underscore that high-fidelity synthetic data is a powerful means to bridge the domain gap in safety-critical applications. The combination of synthetic generation and lightweight adaptation offers a cost-effective, scalable pathway for deploying vision systems in industrial environments where real data is scarce/impractical to obtain. Project Page: https://synspill.vercel.app

Problem

• Detecting rare industrial spills with limited real data
• Improving spill detection using synthetic data generation
• Enhancing vision models for safety-critical industrial applications

Innovation

• Synthetic data generation for spill detection
• Parameter-Efficient Fine-Tuning of VLMs
• Lightweight adaptation for industrial vision systems