[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86152-en":3,"doc-seo-86152-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86152,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","GHOST Geometry-Guided Hallucination of Opaque Surface Textures","Transparent objects challenge depth estimation and 3D reconstruction because they break Lambertian assumptions, producing severe geometric degradation in downstream pipelines. GHOST introduces a geometry-guided preprocessing framework using visual foundation models to convert transparent regions into structurally consistent opaque representations, without retraining downstream models. The approach disentangles masks and physical transparency via TransDINO and TransDecomp, then recovers surface-normal priors with DAF-Net to encode curvature. GeoSemTransNet fuses cues to synthesize texture-rich opaque RGB that preserves the original 3D structure, improving depth and reconstruction accuracy by restoring essential photometric signals.","GHOST: Geometry-Guided Hallucination of Opaque Surface Textures  \nLangxu Zhao * 1 Zuan Gu * 1 Tianhan Gao 1  \narXiv :2607 . 11118v1 [ cs .CV] 13 Jul 2026  \nAbstract  \nTransparent objects pose a fundamental challenge for depth estimation and 3D reconstruction due to their violation of Lambertian assumptions, leading to severe geometry degradation in downstream tasks. To address this, we propose a novel geometry-guided preprocessing framework GHOST that leverages visual foundation models to transform transparent regions into opaque, structurally consistent representations without requiring downstream model retraining. Specifically, our pipeline utilizes (1) TransDINO and  \n(2) TransDecomp to disentangle masks and transparency physical properties, while (3) DAF-Net recovers surface normal priors to encode geometric curvature. Subsequently, (4) GeoSemTransNet integrates these multi-modal cues to synthesize a texture-rich opaque RGB image that preserves the transparent object’s 3D structure. Extensive experiments demonstrate that our method significantly enhances the accuracy of state-of-the-art depth estimation and reconstruction models on transparent objects by restoring essential photometric cues.  \n1. Introduction  \nMainstream depth estimation and 3D reconstruction models fundamentally assume Lambertian diffuse reflection (Kutulakos & Steger, 2008 ; Sajjan et al., 2020) . However, transparent surfaces violate this via light transmission and specular reflection, causing severe geometric distortion. Recent academic efforts primarily pursue two directions: depth completion (Sajjan et al., 2020) and Neural Radiance Fields (NeRF) based reconstruction (Ichnowski et al., 2021) .  \n1 School of Software, Northeastern University, Shenyang, China. Correspondence to: Tianhan Gao \u003C[gaoth@mail.neu.edu.cn](gaoth@mail.neu.edu.cn) >.  \nProceedings of the 43 rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026 . Copyright 2026 by the author(s) .  \n1.1. Existing Solutions and Limitations  \nDepth completion (Zhang & Funkhouser, 2018) uses deep networks to predict surface normals and boundaries, repairing depth via global optimization. However, these postprocessing methods (Zhang et al., 2023) lack universality and efficiency due to their reliance on specific end-to-end training. Obtaining ground-truth depth for transparent objects is costly, requiring specialized hardware or powderspraying (Zhu et al., 2021) . Consequently, these models generalize poorly to unseen materials or lighting. Meanwhile, NeRF-based approaches suffer from low inference efficiency, requiring multi-view inputs and lengthy per-scene training. This forces a trade-off: abandoning versatile SOTA models for narrow, task-specific ones—creating a ”perception silo.”  \n1.2. Universal Preprocessing Mechanism  \nCore challenges include expensive and scarce GT labels, low efficiency, and incompatibility with generic SOTAs. Visual Foundation Models (VFMs) (Awais et al., 2023) offera solution: self-supervised models like DINOv3 (Simoni et al., 2025) capture dense features and structural boundaries of transparent objects despite textural variations. Since downstream models primarily support RGB, modifying only transparent pixels to mimic opaque counterparts can correct inference results. We propose GHOST (Geometryguided Hierarchical Opaque Style Transfer), a universal pre-processor that ”repaints” transparent regions into structurally consistent opaque objects before downstream processing. GHOST uses DINOv3 for region disentanglement, DAF-Net for geometric priors, and GeoSemTransNet for”opaquification.” This resolves generalization issues without retraining downstream models. Our contributions include:  \nTransDINO: A U-Net-based segmentation model injecting DINOv3 features for robust transparent object mask extraction.  \nTransDecomp: A dual-head ViT architecture that disentangles physical properties by separately estimating transparency and foreground values.","cbCaitZtbPDXQusT","https://ap.wps.com/l/cbCaitZtbPDXQusT","pdf",35611560,7,1,13,"English","en",105,"# Introduction\n## Existing Solutions and Limitations\n## Universal Preprocessing Mechanism\n# Related Work\n## Depth Estimation & 3D Reconstruction","[{\"question\":\"What problem does GHOST address in transparent-object depth estimation and 3D reconstruction?\",\"answer\":\"Transparent surfaces violate Lambertian assumptions through light transmission and specular reflection, leading to severe geometric distortion in downstream depth estimation and reconstruction tasks.\"},{\"question\":\"How does GHOST convert transparent regions into usable opaque representations without retraining downstream models?\",\"answer\":\"It preprocesses inputs by disentangling transparent masks and physical properties (TransDINO, TransDecomp), recovering surface-normal priors (DAF-Net), and then synthesizing texture-rich opaque RGB via GeoSemTransNet.\"},{\"question\":\"What role do the three main components TransDINO, TransDecomp, and DAF-Net play?\",\"answer\":\"TransDINO extracts robust transparent-object masks using DINOv3 features; TransDecomp disentangles transparency and foreground physical values; DAF-Net predicts surface normals to provide geometric curvature priors for subsequent texture synthesis.\"}]",1784208943,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"ghost-geometry-guided-hallucination-of-opaque-surface-textures","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/ghost-geometry-guided-hallucination-of-opaque-surface-textures/86152/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does GHOST address in transparent-object depth estimation and 3D reconstruction?","Question",{"text":76,"@type":77},"Transparent surfaces violate Lambertian assumptions through light transmission and specular reflection, leading to severe geometric distortion in downstream depth estimation and reconstruction tasks.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does GHOST convert transparent regions into usable opaque representations without retraining downstream models?",{"text":81,"@type":77},"It preprocesses inputs by disentangling transparent masks and physical properties (TransDINO, TransDecomp), recovering surface-normal priors (DAF-Net), and then synthesizing texture-rich opaque RGB via GeoSemTransNet.",{"name":83,"@type":74,"acceptedAnswer":84},"What role do the three main components TransDINO, TransDecomp, and DAF-Net play?",{"text":85,"@type":77},"TransDINO extracts robust transparent-object masks using DINOv3 features; TransDecomp disentangles transparency and foreground physical values; DAF-Net predicts surface normals to provide geometric curvature priors for subsequent texture synthesis.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]