English

Exploring Model Quantization in GenAI-based Image Inpainting and Detection of Arable Plants

Computer Vision and Pattern Recognition 2025-03-05 v1 Artificial Intelligence

Abstract

Deep learning-based weed control systems often suffer from limited training data diversity and constrained on-board computation, impacting their real-world performance. To overcome these challenges, we propose a framework that leverages Stable Diffusion-based inpainting to augment training data progressively in 10% increments -- up to an additional 200%, thus enhancing both the volume and diversity of samples. Our approach is evaluated on two state-of-the-art object detection models, YOLO11(l) and RT-DETR(l), using the mAP50 metric to assess detection performance. We explore quantization strategies (FP16 and INT8) for both the generative inpainting and detection models to strike a balance between inference speed and accuracy. Deployment of the downstream models on the Jetson Orin Nano demonstrates the practical viability of our framework in resource-constrained environments, ultimately improving detection accuracy and computational efficiency in intelligent weed management systems.

Keywords

Cite

@article{arxiv.2503.02420,
  title  = {Exploring Model Quantization in GenAI-based Image Inpainting and Detection of Arable Plants},
  author = {Sourav Modak and Ahmet Oğuz Saltık and Anthony Stein},
  journal= {arXiv preprint arXiv:2503.02420},
  year   = {2025}
}