Hyper-parameter tuning for text guided image editing

TL;DR

This paper will elaborate the workflow of Forgedit editing stage with examples, and show how to tune the hyper-parameters in an efficient way to obtain ideal editing results.

Abstract

The test-time finetuning text-guided image editing method, Forgedit, is capable of tackling general and complex image editing problems given only the input image itself and the target text prompt. During finetuning stage, using the same set of finetuning hyper-paramters every time for every given image, Forgedit remembers and understands the input image in 30 seconds. During editing stage, the workflow of Forgedit might seem complicated. However, in fact, the editing process of Forgedit is not more complex than previous SOTA Imagic, yet completely solves the overfitting problem of Imagic. In this paper, we will elaborate the workflow of Forgedit editing stage with examples. We will show how to tune the hyper-parameters in an efficient way to obtain ideal editing results.

Figures (2)

• Figure 1: The workflow of Forgedit, the most usual flow of editing process is highlighted in the figure, i.e. simple vector subtraction and default forgetting strategies according to our findings of the disentangle rules of UNet.
• Figure 2: We show the practical workflow of Forgedit, with testing images from EditEval. In most cases, simple vector subtraction would finish the job. For other hard cases, the default forgetting strategies, 'encoderattn' or 'decoderattn' according to editing intention on structrue or appearance, could solve the problems.