方法对比

	多模态图像分类	多模态BERT分类
领域	深度学习	深度学习
方法族	Machine learning	Machine learning
起源年份	2011–2021	2019
提出者	Ngiam et al.; Radford et al. (CLIP)	Kiela, D. et al.; Lu, J. et al.
类型	Multimodal supervised classification	Multimodal transformer classifier
开创性文献	Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., ... & Sutskever, I. (2021). Learning transferable visual models from natural language supervision. Proceedings of the 38th International Conference on Machine Learning (ICML), PMLR 139, 8748–8763.	Kiela, D., Bhooshan, S., Firooz, H., Perez, E., & Testuggine, D. (2019). Supervised multimodal bitransformers for classifying images and text. arXiv preprint arXiv:1909.02950.
别名	multimodal visual classification, image-text classification, vision-language classification, cross-modal image classification	MMBT, multimodal transformer classification, BERT multimodal fusion, vision-language BERT classifier
相关	6	2
摘要	Multimodal image classification extends standard visual classification by incorporating additional modalities — such as text captions, audio, or structured metadata — alongside image features. Separate encoders process each modality, their representations are fused, and a joint classifier assigns the target label. Models such as CLIP demonstrate that image–text alignment enables zero-shot and few-shot image classification at scale.	Multimodal BERT-based classification extends the BERT transformer architecture to jointly encode and classify data from multiple modalities — most commonly text paired with images — by fusing their representations before a final classification head. Introduced prominently around 2019 through models such as MMBT and ViLBERT, it has become a standard approach for tasks where neither text nor image alone carries sufficient information for accurate labeling.