Add-it: Training-Free Object Insertion in Images With Pretrained Diffusion Models

TL;DR

Add-it is a training-free method that uses extended attention in pretrained diffusion models to seamlessly insert objects into images based on text prompts, outperforming supervised methods.

Contribution

It introduces a novel training-free approach that extends attention mechanisms in diffusion models for natural object insertion without fine-tuning.

Findings

Achieves state-of-the-art results on insertion benchmarks.

Preferred in over 80% of human evaluations.

Improves automated metrics for object insertion quality.

Abstract

Adding Object into images based on text instructions is a challenging task in semantic image editing, requiring a balance between preserving the original scene and seamlessly integrating the new object in a fitting location. Despite extensive efforts, existing models often struggle with this balance, particularly with finding a natural location for adding an object in complex scenes. We introduce Add-it, a training-free approach that extends diffusion models' attention mechanisms to incorporate information from three key sources: the scene image, the text prompt, and the generated image itself. Our weighted extended-attention mechanism maintains structural consistency and fine details while ensuring natural object placement. Without task-specific fine-tuning, Add-it achieves state-of-the-art results on both real and generated image insertion benchmarks, including our newly constructed "Additing Affordance Benchmark" for evaluating object placement plausibility, outperforming supervised methods. Human evaluations show that Add-it is preferred in over 80% of cases, and it also demonstrates improvements in various automated metrics.