REVIEW 7 cited by
Designing a Better Asymmetric VQGAN for StableDiffusion
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
StableDiffusion is a revolutionary text-to-image generator that is causing a stir in the world of image generation and editing. Unlike traditional methods that learn a diffusion model in pixel space, StableDiffusion learns a diffusion model in the latent space via a VQGAN, ensuring both efficiency and quality. It not only supports image generation tasks, but also enables image editing for real images, such as image inpainting and local editing. However, we have observed that the vanilla VQGAN used in StableDiffusion leads to significant information loss, causing distortion artifacts even in non-edited image regions. To this end, we propose a new asymmetric VQGAN with two simple designs. Firstly, in addition to the input from the encoder, the decoder contains a conditional branch that incorporates information from task-specific priors, such as the unmasked image region in inpainting. Secondly, the decoder is much heavier than the encoder, allowing for more detailed recovery while only slightly increasing the total inference cost. The training cost of our asymmetric VQGAN is cheap, and we only need to retrain a new asymmetric decoder while keeping the vanilla VQGAN encoder and StableDiffusion unchanged. Our asymmetric VQGAN can be widely used in StableDiffusion-based inpainting and local editing methods. Extensive experiments demonstrate that it can significantly improve the inpainting and editing performance, while maintaining the original text-to-image capability. The code is available at \url{https://github.com/buxiangzhiren/Asymmetric_VQGAN}.
Forward citations
Cited by 7 Pith papers
-
Aligned Stable Inpainting: Mitigating Unwanted Object Insertion and Preserving Color Consistency
ASUKA uses MAE priors and a harmonization VAE decoder to reduce hallucinated objects and color shifts in latent diffusion inpainting.
-
XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations
XR-1 introduces Unified Vision-Motion Codes learned by dual-branch VQ-VAE and applies them in a three-stage training pipeline to outperform prior VLA models on 120+ real-world manipulation tasks across six robot embodiments.
-
CompSlider: Compositional Slider for Disentangled Multiple-Attribute Image Generation
CompSlider learns to synthesize image-conditioning latents from multiple attribute sliders at once, aiming for more disentangled and structure-preserving multi-attribute control in text-to-image generation.
-
LSSGen: Leveraging Latent Space Scaling in Flow and Diffusion for Efficient Text to Image Generation
A latent-space scaling framework that replaces pixel-space upscaling with a trainable latent upsampler and noise compensation, yielding faster high-resolution text-to-image generation.
-
UniLDiff: Unlocking the Power of Diffusion Priors for All-in-One Image Restoration
UniLDiff combines degradation-aware attention fusion with a detail-aware expert decoder to achieve state-of-the-art perceptual quality on unified image restoration benchmarks.
-
PromptArtisan: Multi-instruction Image Editing in Single Pass with Complete Attention Control
PromptArtisan applies multiple mask-prompt edits to an image in one diffusion pass using cross-attention and self-attention control on InstructPix2Pix.
-
Efficient Difficulty-Aware Dynamic Routing for Diffusion-Based Real-World Image Super-Resolution
DDR-SR routes each real-world low-resolution image to one of two diffusion experts based on a high-frequency-loss difficulty score, using a low-compression VAE for hard images and a high-compression VAE for easy image...
Discussion (0). Sign in to comment.