F-SCP: An automatic prompt generation method for specific classes based on visual language pre-training models

Baihong Han, Xiaoyan Jiang, Zhijun Fang, Hamido Fujita, Yongbin Gao

Abstract: Highlights•The prompt greatly affects visual language pre-training models’ classification performance.•Classes with lower accuracy than others and similar classes are specific classes.•We filter out the poorly performing classes for separate prompt generation.•We minimize the impact of specific class prompt on classification of other classes.•Experiments show superiority of our proposed architecture over both CoOp-UC and CoOp-CSC.