Mao, S., Lin, D., Guo, A., & Yiqun, L. (2024). Partclip: How Does Clip Assist Mechanical Part Image Retrieval? 2024 IEEE International Conference on Multimedia and Expo Workshops (ICMEW), 1–5. https://doi.org/10.1109/icmew63481.2024.10645410
Abstract:
CLIP demonstrates impressive performance across several downstream tasks, such as zero-shot image classification. However, these tasks typically involve images from everyday scenarios, and the efficacy of CLIP in domain-specific computer vision tasks associated with the manufacturing industry remains unexplored. This paper first investigates how well CLIP understands the mechanical part images from the manufacturing industrial scenes by conducting a thorough evaluation of its performance in the mechanical part image retrieval task. It turns out that direct employment of CLIP is less effective for this task. At the same time, considering the requirement of this task for deployment on the industry platform in a factory, the large size of the CLIP model presents a practical challenge. Therefore, we explore the knowledge distillation techniques to transfer the knowledge of CLIP into a lighter Efficientnet B1. Our experimental results demonstrate that this CLIP-based knowledge distillation approach can enhance the performance of Efficientnet B1 on mechanical part image retrieval significantly.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the A*STAR - RIE2020 INDUSTRY ALIGNMENT FUND - INDUSTRY COLLABORATION PROJECTS (IAF- ICP)
Grant Reference no. : I2001E0073