Võrdle meetodeid

	Multimodaalne pildiklassifikaator	Pildiklassifikatsiooni peenhäälestus
Valdkond	Süvaõpe	Süvaõpe
Perekond	Machine learning	Machine learning
Tekkeaasta	2011–2021	2010–2014
Looja	Ngiam et al.; Radford et al. (CLIP)	Yosinski, J. et al.; Pan, S. J. & Yang, Q.
Tüüp	Multimodal supervised classification	Transfer learning / fine-tuning
Algallikas	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.	Yosinski, J., Clune, J., Bengio, Y., & Lipson, H. (2014). How transferable are features in deep neural networks? Advances in Neural Information Processing Systems (NeurIPS), 27, 3320–3328.
Rööpnimetused	multimodal visual classification, image-text classification, vision-language classification, cross-modal image classification	fine-tuning for image recognition, transfer learning image classifier, pretrained CNN fine-tuning, domain-specific image classifier
Seotud	6	5
Kokkuvõte	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.	Fine-tuned image classification adapts a large neural network pretrained on a broad image corpus (such as ImageNet) to a specific target domain by continuing training on labeled domain images. This approach achieves strong accuracy with far fewer target-domain samples than training from scratch, making it the dominant paradigm for applied computer vision tasks.