TraDiffusion: Trajectory-Based Training-Free Image Generation

TL;DR

TraDiffusion introduces a training-free, trajectory-based method for controllable image generation that allows users to guide the process using mouse trajectories, enabling precise and natural control over generated content.

Contribution

It presents a novel, training-free approach that uses a distance awareness energy function to guide latent variables based on user-defined trajectories.

Findings

Enables precise control of image regions and attributes

Facilitates manipulation of salient regions and relationships

Works effectively on COCO dataset with natural results

Abstract

In this work, we propose a training-free, trajectory-based controllable T2I approach, termed TraDiffusion. This novel method allows users to effortlessly guide image generation via mouse trajectories. To achieve precise control, we design a distance awareness energy function to effectively guide latent variables, ensuring that the focus of generation is within the areas defined by the trajectory. The energy function encompasses a control function to draw the generation closer to the specified trajectory and a movement function to diminish activity in areas distant from the trajectory. Through extensive experiments and qualitative assessments on the COCO dataset, the results reveal that TraDiffusion facilitates simpler, more natural image control. Moreover, it showcases the ability to manipulate salient regions, attributes, and relationships within the generated images, alongside visual input based on arbitrary or enhanced trajectories.