Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Ledun, Ji, Yatu, Zhuang, Xufei, Yao, Xinying
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.14461
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910219213209600
author Zhang, Ledun
Ji, Yatu
Zhuang, Xufei
Yao, Xinying
author_facet Zhang, Ledun
Ji, Yatu
Zhuang, Xufei
Yao, Xinying
contents Existing object removal tools often rely on manual masks or text prompts, making precise removal difficult for non-expert users in complex scenes and often leading to incomplete removal or unnatural background completion. To address this issue, we present ClickRemoval, an open-source interactive object removal tool built on pretrained Stable Diffusion models and driven solely by user clicks. Without additional training, hand-drawn masks, or text descriptions, ClickRemoval localizes target objects and restores the background through self-attention modulation during denoising. Experiments show that ClickRemoval achieves competitive results across quantitative metrics and user studies. We release a complete software package at https://github.com/zld-make/ClickRemoval under the Apache-2.0 license.
format Preprint
id arxiv_https___arxiv_org_abs_2605_14461
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ClickRemoval: An Interactive Open-Source Tool for Object Removal in Diffusion Models
Zhang, Ledun
Ji, Yatu
Zhuang, Xufei
Yao, Xinying
Computer Vision and Pattern Recognition
Existing object removal tools often rely on manual masks or text prompts, making precise removal difficult for non-expert users in complex scenes and often leading to incomplete removal or unnatural background completion. To address this issue, we present ClickRemoval, an open-source interactive object removal tool built on pretrained Stable Diffusion models and driven solely by user clicks. Without additional training, hand-drawn masks, or text descriptions, ClickRemoval localizes target objects and restores the background through self-attention modulation during denoising. Experiments show that ClickRemoval achieves competitive results across quantitative metrics and user studies. We release a complete software package at https://github.com/zld-make/ClickRemoval under the Apache-2.0 license.
title ClickRemoval: An Interactive Open-Source Tool for Object Removal in Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.14461