Latent Diffusion was proposed in High-Resolution Image Synthesis with Latent Diffusion Models by Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, Björn Ommer.
The abstract of the paper is the following:
By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image data and beyond. Additionally, their formulation allows for a guiding mechanism to control the image generation process without retraining. However, since these models typically operate directly in pixel space, optimization of powerful DMs often consumes hundreds of GPU days and inference is expensive due to sequential evaluations. To enable DM training on limited computational resources while retaining their quality and flexibility, we apply them in the latent space of powerful pretrained autoencoders. In contrast to previous work, training diffusion models on such a representation allows for the first time to reach a near-optimal point between complexity reduction and detail preservation, greatly boosting visual fidelity. By introducing cross-attention layers into the model architecture, we turn diffusion models into powerful and flexible generators for general conditioning inputs such as text or bounding boxes and high-resolution synthesis becomes possible in a convolutional manner. Our latent diffusion models (LDMs) achieve a new state of the art for image inpainting and highly competitive performance on various tasks, including unconditional image generation, semantic scene synthesis, and super-resolution, while significantly reducing computational requirements compared to pixel-based DMs.
The original codebase can be found here.
class diffusers.LDMTextToImagePipeline< source >
( vqvae: typing.Union[diffusers.models.vae.VQModel, diffusers.models.vae.AutoencoderKL] bert: PreTrainedModel tokenizer: PreTrainedTokenizer unet: typing.Union[diffusers.models.unet_2d.UNet2DModel, diffusers.models.unet_2d_condition.UNet2DConditionModel] scheduler: typing.Union[diffusers.schedulers.scheduling_ddim.DDIMScheduler, diffusers.schedulers.scheduling_pndm.PNDMScheduler, diffusers.schedulers.scheduling_lms_discrete.LMSDiscreteScheduler] )
- vqvae (VQModel) — Vector-quantized (VQ) Model to encode and decode images to and from latent representations.
LDMBertModel) — Text-encoder model based on BERT architecture.
transformers.BertTokenizer) — Tokenizer of class BertTokenizer.
- unet (UNet2DConditionModel) — Conditional U-Net architecture to denoise the encoded image latents.
scheduler (SchedulerMixin) —
A scheduler to be used in combination with
unetto denoise the encoded image latens. Can be one of DDIMScheduler, LMSDiscreteScheduler, or PNDMScheduler.
This model inherits from DiffusionPipeline. Check the superclass documentation for the generic methods the library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
__call__< source >
prompt: typing.Union[str, typing.List[str]]
height: typing.Optional[int] = 256
width: typing.Optional[int] = 256
num_inference_steps: typing.Optional[int] = 50
guidance_scale: typing.Optional[float] = 1.0
eta: typing.Optional[float] = 0.0
generator: typing.Optional[torch._C.Generator] = None
output_type: typing.Optional[str] = 'pil'
return_dict: bool = True
List[str]) — The prompt or prompts to guide the image generation.
int, optional, defaults to 256) — The height in pixels of the generated image.
int, optional, defaults to 256) — The width in pixels of the generated image.
int, optional, defaults to 50) — The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference.
float, optional, defaults to 1.0) — Guidance scale as defined in Classifier-Free Diffusion Guidance.
guidance_scaleis defined as
wof equation 2. of Imagen Paper. Guidance scale is enabled by setting
guidance_scale > 1. Higher guidance scale encourages to generate images that are closely linked to the text
promptat the, usually at the expense of lower image quality.
torch.Generator, optional) — A torch generator to make generation deterministic.
str, optional, defaults to
"pil") — The output format of the generate image. Choose between PIL:
bool, optional) — Whether or not to return a ImagePipelineOutput instead of a plain tuple.
return_dict is True, otherwise a `tuple. When returning a tuple, the first element is a list with the