Jämför metoder

	Latent Diffusion Models	Swin Transformer	Vision Transformer
Ämnesområde	Djupinlärning	Djupinlärning	Djupinlärning
Familj	Machine learning	Machine learning	Machine learning
Ursprungsår	2022	2021	2021
Upphovsperson	Robin Rombach	Ze Liu	Dosovitskiy, A. et al.
Typ	Neural network architecture	Neural network architecture	Transformer architecture for images (self-attention over patches)
Ursprungskälla	Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022). High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10684-10695).	Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021). Swin Transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 10012-10022).	Dosovitskiy, A. et al. (2021). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. ICLR.
Alias	LDM, Stable Diffusion, Latent Diffusion	Swin, Hierarchical Vision Transformer	Görsel Transformer (ViT), görsel transformer, ViT, patch transformer for images
Närliggande	4	4	5
Sammanfattning	Latent Diffusion Models (LDMs) are a generative approach introduced by Rombach et al. in 2022 that performs the diffusion process in a compressed latent space rather than pixel space, enabling efficient high-resolution image synthesis. By compressing images into a low-dimensional latent representation using a variational autoencoder, diffusion becomes computationally tractable while maintaining visual quality.	The Swin Transformer is a hierarchical vision transformer introduced by Liu et al. in 2021 that uses shifted window attention to achieve computational efficiency while maintaining strong performance on computer vision tasks. Unlike the original Vision Transformer which applies global self-attention, Swin uses local window-based attention with periodic shifting to balance expressiveness and efficiency.	The Vision Transformer (ViT), introduced by Dosovitskiy and colleagues in 2021, splits an image into fixed-size patches, treats those patches as a sequence, and applies the Transformer self-attention mechanism to image classification. Given enough training data, it surpasses convolutional neural networks (CNNs).