Curved Diffusion: A Generative Model With Optical Geometry Control

TL;DR

This paper presents a diffusion model framework that incorporates optical geometry control, allowing for diverse visual effects like fish-eye and panoramic views by manipulating lens curvature during image generation.

Contribution

It introduces a novel per-pixel coordinate conditioning approach to integrate camera lens geometry into diffusion models, enabling explicit control over optical effects.

Findings

Controlled curvature manipulation produces diverse visual effects

Single model generates fish-eye, panoramic, and spherical images

Optical geometry integration enhances image realism and diversity

Abstract

State-of-the-art diffusion models can generate highly realistic images based on various conditioning like text, segmentation, and depth. However, an essential aspect often overlooked is the specific camera geometry used during image capture. The influence of different optical systems on the final scene appearance is frequently overlooked. This study introduces a framework that intimately integrates a text-to-image diffusion model with the particular lens geometry used in image rendering. Our method is based on a per-pixel coordinate conditioning method, enabling the control over the rendering geometry. Notably, we demonstrate the manipulation of curvature properties, achieving diverse visual effects, such as fish-eye, panoramic views, and spherical texturing using a single diffusion model.