VQ Diffusion
Last updated
Last updated
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 (VQ-Diffusion) model for text-to-image generation. This method is based on a vector quantized variational autoencoder (VQ-VAE) whose latent space is modeled by a conditional variant of the recently developed Denoising Diffusion Probabilistic Model (DDPM). We find that this latent-space method is well-suited for text-to-image generation tasks because it not only eliminates the unidirectional bias with existing methods but also allows us to incorporate a mask-and-replace diffusion strategy to avoid the accumulation of errors, which is a serious problem with existing methods. Our experiments show that the VQ-Diffusion produces significantly better text-to-image generation results when compared with conventional autoregressive (AR) models with similar numbers of parameters. Compared with previous GAN-based text-to-image methods, our VQ-Diffusion 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 text-to-image generation time increases linearly with the output image resolution and hence is quite time consuming even for normal size images. The VQ-Diffusion allows us to achieve a better trade-off between quality and speed. Our experiments indicate that the VQ-Diffusion 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 .
Make sure to check out the Schedulers to learn how to explore the tradeoff between scheduler speed and quality, and see the section to learn how to efficiently load the same components into multiple pipelines.
( vqvae: VQModeltext_encoder: CLIPTextModeltokenizer: CLIPTokenizertransformer: Transformer2DModelscheduler: VQDiffusionSchedulerlearned_classifier_free_sampling_embeddings: LearnedClassifierFreeSamplingEmbeddings )
Parameters
vqvae () — Vector Quantized Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
text_encoder (CLIPTextModel) — Frozen text-encoder ().
tokenizer (CLIPTokenizer) — A CLIPTokenizer to tokenize text.
transformer () — A conditional Transformer2DModel to denoise the encoded image latents.
scheduler () — A scheduler to be used in combination with transformer to denoise the encoded image latents.
Pipeline for text-to-image generation using VQ Diffusion.
__call__
Parameters
prompt (str or List[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 text prompt at the expense of lower image quality. Guidance scale is enabled when guidance_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 most truncation_rate. The lowest probabilities that would increase the cumulative probability above truncation_rate are set to zero.
num_images_per_prompt (int, optional, defaults to 1) — The number of images to generate per prompt.
latents (torch.FloatTensor of shape (batch), optional) — Pre-generated 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 between PIL.Image or np.array.
callback (Callable, optional) — A function that calls every callback_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 the callback function is called. If not specified, the callback is called at every step.
Returns
The call function to the pipeline for generation.
truncate
( log_p_x_0: FloatTensortruncation_rate: float )
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.
( images: typing.Union[typing.List[PIL.Image.Image], numpy.ndarray] )
Parameters
images (List[PIL.Image.Image] or np.ndarray) — List of denoised PIL images of length batch_size or NumPy array of shape (batch_size, height, width, num_channels).
Output class for image pipelines.
This model inherits from . Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.).
( prompt: typing.Union[str, typing.List[str]]num_inference_steps: int = 100guidance_scale: float = 5.0truncation_rate: float = 1.0num_images_per_prompt: int = 1generator: typing.Union[torch._C.Generator, typing.List[torch._C.Generator], NoneType] = Nonelatents: typing.Optional[torch.FloatTensor] = Noneoutput_type: typing.Optional[str] = 'pil'return_dict: bool = Truecallback: typing.Union[typing.Callable[[int, int, torch.FloatTensor], NoneType], NoneType] = Nonecallback_steps: int = 1 ) → or tuple
generator (torch.Generator, optional) — A to make generation deterministic.
return_dict (bool, optional, defaults to True) — Whether or not to return a instead of a plain tuple.
or tuple
If return_dict is True, is returned, otherwise a tuple is returned where the first element is a list with the generated images.