SmartBrush: Text and Shape Guided Object Inpainting with Diffusion Model

By Shaoan Xie, Zhifei Zhang, Zhe L. Lin, Tobias Hinz, Kun Zhang. Computer Vision and Pattern Recognition, 2022

379 citations recorded by the source corpus.

Generic image inpainting aims to complete a corrupted image by borrowing surrounding information, which barely generates novel content. By contrast, multi-modal inpainting provides more flexible and useful controls on the inpainted content, e.g., a text prompt can be used to describe an object with richer attributes, and a mask can be used to constrain the shape of the inpainted object rather than being only considered as a missing area. We propose a new diffusion-based model named SmartBrush for completing a missing region with an object using both text and shape-guidance. While previous work such as DALLE-2 and Stable Diffusion can do text-guided inapinting they do not support shape guidance and tend to modify background texture surrounding the generated object. Our model incorporates both text and shape guidance with precision control. To preserve the background better, we propose a novel training and sampling strategy by augmenting the diffusion U-net with object-mask prediction. Lastly, we introduce a multi-task training strategy by jointly training inpainting with text-to-image generation to leverage more training data. We conduct extensive experiments showing that our model outperforms all baselines in terms of visual quality, mask controllability, and background preservation.

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/95ca7eb71b2aea74d7bbe23252e41c03070b807d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

AnyDoor: Zero-shot Object-level Image Customization (2023, neutral): “Other works [53, 59] explore re-generating a local image region with the guidance of text prompts.”
Improved Distribution Matching Distillation for Fast Image Synthesis (2024, background): “Diffusion models have achieved unprecedented quality in visual generation tasks [1–8].”
VideoComposer: Compositional Video Synthesis with Motion Controllability (2023, methodology): “Most existing methods typically achieve controllable generation mainly by introducing new conditions, such as segmentation maps [44, 59], inpainting masks [66] or sketches [34, 72], in addition to texts.”
Emu Edit: Precise Image Editing via Recognition and Generation Tasks (2023, background): “Imagen Editor [23] and SmartBrush [24] finetune the text-to-image model to be conditioned on both the input image and mask.”
Composer: Creative and Controllable Image Synthesis with Composable Conditions (2023, methodology): “ai, 2022), and inpainting masks (Xie et al., 2022; Wang et al., 2022a), or by finetuning the pretrained models on a few subject-specific data (Gal et al.”