Unconditional Latent Diffusion
Unconditional 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 from the paper is:
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 at CompVis/latent-diffusion.
Make sure to check out the Schedulers guide to learn how to explore the tradeoff between scheduler speed and quality, and see the reuse components across pipelines section to learn how to efficiently load the same components into multiple pipelines.
LDMPipeline
class diffusers.LDMPipeline
< source >( vqvae: VQModel unet: UNet2DModel scheduler: DDIMScheduler )
Parameters
- vqvae (VQModel) — Vector-quantized (VQ) model to encode and decode images to and from latent representations.
- unet (UNet2DModel) —
A
UNet2DModel
to denoise the encoded image latents. - scheduler (SchedulerMixin) —
DDIMScheduler is used in combination with
unet
to denoise the encoded image latents.
Pipeline for unconditional image generation using latent diffusion.
This model inherits from DiffusionPipeline. Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.).
__call__
< source >( batch_size: int = 1 generator: typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None eta: float = 0.0 num_inference_steps: int = 50 output_type: typing.Optional[str] = 'pil' return_dict: bool = True **kwargs ) → ImagePipelineOutput or tuple
Parameters
- batch_size (
int
, optional, defaults to 1) — Number of images to generate. - generator (
torch.Generator
, optional) — Atorch.Generator
to make generation deterministic. - num_inference_steps (
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. - output_type (
str
, optional, defaults to"pil"
) — The output format of the generated image. Choose betweenPIL.Image
ornp.array
. - return_dict (
bool
, optional, defaults toTrue
) — Whether or not to return a ImagePipelineOutput instead of a plain tuple.
Returns
ImagePipelineOutput or tuple
If return_dict
is True
, ImagePipelineOutput is returned, otherwise a tuple
is
returned where the first element is a list with the generated images
The call function to the pipeline for generation.
ImagePipelineOutput
class diffusers.ImagePipelineOutput
< source >( images: typing.Union[typing.List[PIL.Image.Image], numpy.ndarray] )
Output class for image pipelines.