Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Kejie, Song, Xuemeng, Liu, Meng, Yuan, Jin, Guan, Weili
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2410.10496
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916485057740800
author Wang, Kejie
Song, Xuemeng
Liu, Meng
Yuan, Jin
Guan, Weili
author_facet Wang, Kejie
Song, Xuemeng
Liu, Meng
Yuan, Jin
Guan, Weili
contents Text-to-image diffusion models have demonstrated remarkable progress in synthesizing high-quality images from text prompts, which boosts researches on prompt-based image editing that edits a source image according to a target prompt. Despite their advances, existing methods still encounter three key issues: 1) limited capacity of the text prompt in guiding target image generation, 2) insufficient mining of word-to-patch and patch-to-patch relationships for grounding editing areas, and 3) unified editing strength for all regions during each denoising step. To address these issues, we present a Vision-guided and Mask-enhanced Adaptive Editing (ViMAEdit) method with three key novel designs. First, we propose to leverage image embeddings as explicit guidance to enhance the conventional textual prompt-based denoising process, where a CLIP-based target image embedding estimation strategy is introduced. Second, we devise a self-attention-guided iterative editing area grounding strategy, which iteratively exploits patch-to-patch relationships conveyed by self-attention maps to refine those word-to-patch relationships contained in cross-attention maps. Last, we present a spatially adaptive variance-guided sampling, which highlights sampling variances for critical image regions to promote the editing capability. Experimental results demonstrate the superior editing capacity of ViMAEdit over all existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10496
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Vision-guided and Mask-enhanced Adaptive Denoising for Prompt-based Image Editing
Wang, Kejie
Song, Xuemeng
Liu, Meng
Yuan, Jin
Guan, Weili
Computer Vision and Pattern Recognition
Text-to-image diffusion models have demonstrated remarkable progress in synthesizing high-quality images from text prompts, which boosts researches on prompt-based image editing that edits a source image according to a target prompt. Despite their advances, existing methods still encounter three key issues: 1) limited capacity of the text prompt in guiding target image generation, 2) insufficient mining of word-to-patch and patch-to-patch relationships for grounding editing areas, and 3) unified editing strength for all regions during each denoising step. To address these issues, we present a Vision-guided and Mask-enhanced Adaptive Editing (ViMAEdit) method with three key novel designs. First, we propose to leverage image embeddings as explicit guidance to enhance the conventional textual prompt-based denoising process, where a CLIP-based target image embedding estimation strategy is introduced. Second, we devise a self-attention-guided iterative editing area grounding strategy, which iteratively exploits patch-to-patch relationships conveyed by self-attention maps to refine those word-to-patch relationships contained in cross-attention maps. Last, we present a spatially adaptive variance-guided sampling, which highlights sampling variances for critical image regions to promote the editing capability. Experimental results demonstrate the superior editing capacity of ViMAEdit over all existing methods.
title Vision-guided and Mask-enhanced Adaptive Denoising for Prompt-based Image Editing
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.10496