FaithFill: Faithful Inpainting for Object Completion Using a Single Reference Image

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Mallick, Rupayan, Abdalla, Amr, Bargal, Sarah Adel
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914832514547712
author Mallick, Rupayan
Abdalla, Amr
Bargal, Sarah Adel
author_facet Mallick, Rupayan
Abdalla, Amr
Bargal, Sarah Adel
contents We present FaithFill, a diffusion-based inpainting object completion approach for realistic generation of missing object parts. Typically, multiple reference images are needed to achieve such realistic generation, otherwise the generation would not faithfully preserve shape, texture, color, and background. In this work, we propose a pipeline that utilizes only a single input reference image -having varying lighting, background, object pose, and/or viewpoint. The singular reference image is used to generate multiple views of the object to be inpainted. We demonstrate that FaithFill produces faithful generation of the object's missing parts, together with background/scene preservation, from a single reference image. This is demonstrated through standard similarity metrics, human judgement, and GPT evaluation. Our results are presented on the DreamBooth dataset, and a novel proposed dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2406_07865
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FaithFill: Faithful Inpainting for Object Completion Using a Single Reference Image
Mallick, Rupayan
Abdalla, Amr
Bargal, Sarah Adel
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
We present FaithFill, a diffusion-based inpainting object completion approach for realistic generation of missing object parts. Typically, multiple reference images are needed to achieve such realistic generation, otherwise the generation would not faithfully preserve shape, texture, color, and background. In this work, we propose a pipeline that utilizes only a single input reference image -having varying lighting, background, object pose, and/or viewpoint. The singular reference image is used to generate multiple views of the object to be inpainted. We demonstrate that FaithFill produces faithful generation of the object's missing parts, together with background/scene preservation, from a single reference image. This is demonstrated through standard similarity metrics, human judgement, and GPT evaluation. Our results are presented on the DreamBooth dataset, and a novel proposed dataset.
title FaithFill: Faithful Inpainting for Object Completion Using a Single Reference Image
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2406.07865