Novel Object Synthesis via Adaptive Text-Image Harmony

NeurIPS 2024

Zeren Xiong¹, Zedong Zhang¹, Zikun Chen¹, Shuo Chen²,
Xiang Li³, Gan Sun⁴, Jian Yang¹, Jun Li^1*

¹Nanjing University of Science and Technology ²RIKEN
³Nankai University ⁴South China University of Technology

Abstract

In this paper, we study an object synthesis task that combines an object text with an object image to create a new object image. However, most diffusion models struggle with this task,i.e., often generating an object that predominantly reflects either the text or the image due to an imbalance between their inputs. To address this issue, we propose a simple yet effective method called Adaptive Text-Image Harmony (ATIH) to generate novel and surprising objects. First, we introduce a scale factor and an injection step to balance text and image features in cross-attention and to preserve image information in self-attention during the text-image inversion diffusion process, respectively. Second, to better integrate object text and image, we design a balanced loss function with a noise parameter, ensuring both optimal editability and fidelity of the object image. Third, to adaptively adjust these parameters, we present a novel similarity score function that not only maximizes the similarities between the generated object image and the input text/image but also balances these similarities to harmonize text and image integration.