VQ Diffusion
Vector Quantized Diffusion Model for TexttoImage Synthesis is by Shuyang Gu, Dong Chen, Jianmin Bao, Fang Wen, Bo Zhang, Dongdong Chen, Lu Yuan, Baining Guo.
The abstract from the paper is:
We present the vector quantized diffusion (VQDiffusion) model for texttoimage generation. This method is based on a vector quantized variational autoencoder (VQVAE) whose latent space is modeled by a conditional variant of the recently developed Denoising Diffusion Probabilistic Model (DDPM). We find that this latentspace method is wellsuited for texttoimage generation tasks because it not only eliminates the unidirectional bias with existing methods but also allows us to incorporate a maskandreplace diffusion strategy to avoid the accumulation of errors, which is a serious problem with existing methods. Our experiments show that the VQDiffusion produces significantly better texttoimage generation results when compared with conventional autoregressive (AR) models with similar numbers of parameters. Compared with previous GANbased texttoimage methods, our VQDiffusion can handle more complex scenes and improve the synthesized image quality by a large margin. Finally, we show that the image generation computation in our method can be made highly efficient by reparameterization. With traditional AR methods, the texttoimage generation time increases linearly with the output image resolution and hence is quite time consuming even for normal size images. The VQDiffusion allows us to achieve a better tradeoff between quality and speed. Our experiments indicate that the VQDiffusion model with the reparameterization is fifteen times faster than traditional AR methods while achieving a better image quality.
The original codebase can be found at microsoft/VQDiffusion.
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.
VQDiffusionPipeline
class diffusers.VQDiffusionPipeline
< source >( vqvae: VQModel text_encoder: CLIPTextModel tokenizer: CLIPTokenizer transformer: Transformer2DModel scheduler: VQDiffusionScheduler learned_classifier_free_sampling_embeddings: LearnedClassifierFreeSamplingEmbeddings )
Parameters
 vqvae (VQModel) — Vector Quantized Variational AutoEncoder (VAE) model to encode and decode images to and from latent representations.

text_encoder (
CLIPTextModel
) — Frozen textencoder (clipvitbasepatch32). 
tokenizer (
CLIPTokenizer
) — ACLIPTokenizer
to tokenize text. 
transformer (Transformer2DModel) —
A conditional
Transformer2DModel
to denoise the encoded image latents. 
scheduler (VQDiffusionScheduler) —
A scheduler to be used in combination with
transformer
to denoise the encoded image latents.
Pipeline for texttoimage generation using VQ 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 >(
prompt: typing.Union[str, typing.List[str]]
num_inference_steps: int = 100
guidance_scale: float = 5.0
truncation_rate: float = 1.0
num_images_per_prompt: int = 1
generator: typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = None
latents: typing.Optional[torch.FloatTensor] = None
output_type: typing.Optional[str] = 'pil'
return_dict: bool = True
callback: typing.Union[typing.Callable[[int, int, torch.FloatTensor], NoneType], NoneType] = None
callback_steps: int = 1
)
→
ImagePipelineOutput or tuple
Parameters

prompt (
str
orList[str]
) — The prompt or prompts to guide image generation. 
num_inference_steps (
int
, optional, defaults to 100) — The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference. 
guidance_scale (
float
, optional, defaults to 7.5) — A higher guidance scale value encourages the model to generate images closely linked to the textprompt
at the expense of lower image quality. Guidance scale is enabled whenguidance_scale > 1
. 
truncation_rate (
float
, optional, defaults to 1.0 (equivalent to no truncation)) — Used to “truncate” the predicted classes for x_0 such that the cumulative probability for a pixel is at mosttruncation_rate
. The lowest probabilities that would increase the cumulative probability abovetruncation_rate
are set to zero. 
num_images_per_prompt (
int
, optional, defaults to 1) — The number of images to generate per prompt. 
generator (
torch.Generator
, optional) — Atorch.Generator
to make generation deterministic. 
latents (
torch.FloatTensor
of shape (batch), optional) — Pregenerated noisy latents sampled from a Gaussian distribution, to be used as inputs for image generation. Must be valid embedding indices.If not provided, a latents tensor will be generated of completely masked latent pixels. 
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. 
callback (
Callable
, optional) — A function that calls everycallback_steps
steps during inference. The function is called with the following arguments:callback(step: int, timestep: int, latents: torch.FloatTensor)
. 
callback_steps (
int
, optional, defaults to 1) — The frequency at which thecallback
function is called. If not specified, the callback is called at every step.
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.
Truncates log_p_x_0
such that for each column vector, the total cumulative probability is truncation_rate
The lowest probabilities that would increase the cumulative probability above truncation_rate
are set to
zero.
ImagePipelineOutput
class diffusers.ImagePipelineOutput
< source >( images: typing.Union[typing.List[PIL.Image.Image], numpy.ndarray] )
Output class for image pipelines.