Reflecting Reality: Enabling Diffusion Models to Produce Faithful Mirror Reflections

We tackle the problem of generating highly realistic and plausible mirror reflections using diffusion-based generative models. We formulate this problem as an image inpainting task, allowing for more user control over the placement of mirrors during the generation process. To enable this, we create...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dhiman, Ankit, Shah, Manan, Parihar, Rishubh, Bhalgat, Yash, Boregowda, Lokesh R, Babu, R Venkatesh
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Dhiman, Ankit
Shah, Manan
Parihar, Rishubh
Bhalgat, Yash
Boregowda, Lokesh R
Babu, R Venkatesh
description We tackle the problem of generating highly realistic and plausible mirror reflections using diffusion-based generative models. We formulate this problem as an image inpainting task, allowing for more user control over the placement of mirrors during the generation process. To enable this, we create SynMirror, a large-scale dataset of diverse synthetic scenes with objects placed in front of mirrors. SynMirror contains around 198K samples rendered from 66K unique 3D objects, along with their associated depth maps, normal maps and instance-wise segmentation masks, to capture relevant geometric properties of the scene. Using this dataset, we propose a novel depth-conditioned inpainting method called MirrorFusion, which generates high-quality geometrically consistent and photo-realistic mirror reflections given an input image and a mask depicting the mirror region. MirrorFusion outperforms state-of-the-art methods on SynMirror, as demonstrated by extensive quantitative and qualitative analysis. To the best of our knowledge, we are the first to successfully tackle the challenging problem of generating controlled and faithful mirror reflections of an object in a scene using diffusion based models. SynMirror and MirrorFusion open up new avenues for image editing and augmented reality applications for practitioners and researchers alike.
doi_str_mv 10.48550/arxiv.2409.14677
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2409_14677</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2409_14677</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2409_146773</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjGw1DM0MTM352QICUpNy0lNLsnMS1cISk3MySyptFJwzUtMygGJuGSmpZUWZ-bnKfjmp6TmFCuU5CsEFOWnlCanKrglZpZkpJXmKPhmFhXlFynADMrPK-ZhYE1LzClO5YXS3Azybq4hzh66YPvjC4oycxOLKuNB7ogHu8OYsAoAFJw-Qw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Reflecting Reality: Enabling Diffusion Models to Produce Faithful Mirror Reflections</title><source>arXiv.org</source><creator>Dhiman, Ankit ; Shah, Manan ; Parihar, Rishubh ; Bhalgat, Yash ; Boregowda, Lokesh R ; Babu, R Venkatesh</creator><creatorcontrib>Dhiman, Ankit ; Shah, Manan ; Parihar, Rishubh ; Bhalgat, Yash ; Boregowda, Lokesh R ; Babu, R Venkatesh</creatorcontrib><description>We tackle the problem of generating highly realistic and plausible mirror reflections using diffusion-based generative models. We formulate this problem as an image inpainting task, allowing for more user control over the placement of mirrors during the generation process. To enable this, we create SynMirror, a large-scale dataset of diverse synthetic scenes with objects placed in front of mirrors. SynMirror contains around 198K samples rendered from 66K unique 3D objects, along with their associated depth maps, normal maps and instance-wise segmentation masks, to capture relevant geometric properties of the scene. Using this dataset, we propose a novel depth-conditioned inpainting method called MirrorFusion, which generates high-quality geometrically consistent and photo-realistic mirror reflections given an input image and a mask depicting the mirror region. MirrorFusion outperforms state-of-the-art methods on SynMirror, as demonstrated by extensive quantitative and qualitative analysis. To the best of our knowledge, we are the first to successfully tackle the challenging problem of generating controlled and faithful mirror reflections of an object in a scene using diffusion based models. SynMirror and MirrorFusion open up new avenues for image editing and augmented reality applications for practitioners and researchers alike.</description><identifier>DOI: 10.48550/arxiv.2409.14677</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2024-09</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2409.14677$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2409.14677$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Dhiman, Ankit</creatorcontrib><creatorcontrib>Shah, Manan</creatorcontrib><creatorcontrib>Parihar, Rishubh</creatorcontrib><creatorcontrib>Bhalgat, Yash</creatorcontrib><creatorcontrib>Boregowda, Lokesh R</creatorcontrib><creatorcontrib>Babu, R Venkatesh</creatorcontrib><title>Reflecting Reality: Enabling Diffusion Models to Produce Faithful Mirror Reflections</title><description>We tackle the problem of generating highly realistic and plausible mirror reflections using diffusion-based generative models. We formulate this problem as an image inpainting task, allowing for more user control over the placement of mirrors during the generation process. To enable this, we create SynMirror, a large-scale dataset of diverse synthetic scenes with objects placed in front of mirrors. SynMirror contains around 198K samples rendered from 66K unique 3D objects, along with their associated depth maps, normal maps and instance-wise segmentation masks, to capture relevant geometric properties of the scene. Using this dataset, we propose a novel depth-conditioned inpainting method called MirrorFusion, which generates high-quality geometrically consistent and photo-realistic mirror reflections given an input image and a mask depicting the mirror region. MirrorFusion outperforms state-of-the-art methods on SynMirror, as demonstrated by extensive quantitative and qualitative analysis. To the best of our knowledge, we are the first to successfully tackle the challenging problem of generating controlled and faithful mirror reflections of an object in a scene using diffusion based models. SynMirror and MirrorFusion open up new avenues for image editing and augmented reality applications for practitioners and researchers alike.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjGw1DM0MTM352QICUpNy0lNLsnMS1cISk3MySyptFJwzUtMygGJuGSmpZUWZ-bnKfjmp6TmFCuU5CsEFOWnlCanKrglZpZkpJXmKPhmFhXlFynADMrPK-ZhYE1LzClO5YXS3Azybq4hzh66YPvjC4oycxOLKuNB7ogHu8OYsAoAFJw-Qw</recordid><startdate>20240922</startdate><enddate>20240922</enddate><creator>Dhiman, Ankit</creator><creator>Shah, Manan</creator><creator>Parihar, Rishubh</creator><creator>Bhalgat, Yash</creator><creator>Boregowda, Lokesh R</creator><creator>Babu, R Venkatesh</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240922</creationdate><title>Reflecting Reality: Enabling Diffusion Models to Produce Faithful Mirror Reflections</title><author>Dhiman, Ankit ; Shah, Manan ; Parihar, Rishubh ; Bhalgat, Yash ; Boregowda, Lokesh R ; Babu, R Venkatesh</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2409_146773</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Dhiman, Ankit</creatorcontrib><creatorcontrib>Shah, Manan</creatorcontrib><creatorcontrib>Parihar, Rishubh</creatorcontrib><creatorcontrib>Bhalgat, Yash</creatorcontrib><creatorcontrib>Boregowda, Lokesh R</creatorcontrib><creatorcontrib>Babu, R Venkatesh</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Dhiman, Ankit</au><au>Shah, Manan</au><au>Parihar, Rishubh</au><au>Bhalgat, Yash</au><au>Boregowda, Lokesh R</au><au>Babu, R Venkatesh</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Reflecting Reality: Enabling Diffusion Models to Produce Faithful Mirror Reflections</atitle><date>2024-09-22</date><risdate>2024</risdate><abstract>We tackle the problem of generating highly realistic and plausible mirror reflections using diffusion-based generative models. We formulate this problem as an image inpainting task, allowing for more user control over the placement of mirrors during the generation process. To enable this, we create SynMirror, a large-scale dataset of diverse synthetic scenes with objects placed in front of mirrors. SynMirror contains around 198K samples rendered from 66K unique 3D objects, along with their associated depth maps, normal maps and instance-wise segmentation masks, to capture relevant geometric properties of the scene. Using this dataset, we propose a novel depth-conditioned inpainting method called MirrorFusion, which generates high-quality geometrically consistent and photo-realistic mirror reflections given an input image and a mask depicting the mirror region. MirrorFusion outperforms state-of-the-art methods on SynMirror, as demonstrated by extensive quantitative and qualitative analysis. To the best of our knowledge, we are the first to successfully tackle the challenging problem of generating controlled and faithful mirror reflections of an object in a scene using diffusion based models. SynMirror and MirrorFusion open up new avenues for image editing and augmented reality applications for practitioners and researchers alike.</abstract><doi>10.48550/arxiv.2409.14677</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2409.14677
ispartof
issn
language eng
recordid cdi_arxiv_primary_2409_14677
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title Reflecting Reality: Enabling Diffusion Models to Produce Faithful Mirror Reflections
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-30T18%3A13%3A18IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Reflecting%20Reality:%20Enabling%20Diffusion%20Models%20to%20Produce%20Faithful%20Mirror%20Reflections&rft.au=Dhiman,%20Ankit&rft.date=2024-09-22&rft_id=info:doi/10.48550/arxiv.2409.14677&rft_dat=%3Carxiv_GOX%3E2409_14677%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true