Vision Transformers versus convolutional neural networks for fine-grained orchid genus identification in a species-rich, data-poor flora: a controlled benchmark on the Orchidaceae of New Guinea
研究使用自监督视觉变换器和嵌入检索方法解决物种丰富但数据稀缺的兰花属识别问题,对比了多种预训练模型,发现DINOv2表现最佳。
0 citationsRead paper