{"paper":{"title":"DealMaTe: Multi-Dimensional Material Transfer via Diffusion Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"DealMaTe transfers materials across objects using depth, normal, and lighting images in a text-free diffusion framework.","cross_cats":["cs.CV"],"primary_cat":"cs.GR","authors_text":"Jie Guo, Nisha Huang, Tong-Yee Lee, Xiu Li, Yizhou Lin, Zitong Yu","submitted_at":"2026-05-15T07:06:39Z","abstract_excerpt":"Recently, diffusion-based material transfer methods rely on image fine-tuning or complex architectures with auxiliary networks but face challenges such as text dependency, additional computational costs, and feature misalignment. To address these limitations, we propose \\textbf{DealMaTe}, using \\underline{\\textbf{de}}pth, norm\\underline{\\textbf{a}}l, and \\underline{\\textbf{l}}ighting images for \\underline{\\textbf{ma}}terial \\underline{\\textbf{t}}ransf\\underline{\\textbf{e}}r. DealMaTe is a simplified diffusion framework that eliminates text guidance and reference networks. We design a lightweig"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Extensive experiments covering a wide variety of objects and lighting conditions consistently demonstrate that DealMaTe achieves remarkable high-fidelity material transfer under arbitrary input materials.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The lightweight 3D information injection method (Multi-Dim 3D Shader LoRA) enables compatible control conditions and achieves harmonious and stable results without modifying the base model weights.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DealMaTe proposes a simplified diffusion framework for material transfer that injects multi-dimensional 3D conditions via Multi-Dim 3D Shader LoRA and Shader Causal Mutual Attention with KV caching.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"DealMaTe transfers materials across objects using depth, normal, and lighting images in a text-free diffusion framework.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"05bb6a8871f272dabfd40214a2901fac0ea72ba10bbbf68335379cdf2a65ae51"},"source":{"id":"2605.15681","kind":"arxiv","version":1},"verdict":{"id":"bdaf239f-dca6-4427-a212-e394b64cce37","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T19:08:34.414842Z","strongest_claim":"Extensive experiments covering a wide variety of objects and lighting conditions consistently demonstrate that DealMaTe achieves remarkable high-fidelity material transfer under arbitrary input materials.","one_line_summary":"DealMaTe proposes a simplified diffusion framework for material transfer that injects multi-dimensional 3D conditions via Multi-Dim 3D Shader LoRA and Shader Causal Mutual Attention with KV caching.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The lightweight 3D information injection method (Multi-Dim 3D Shader LoRA) enables compatible control conditions and achieves harmonious and stable results without modifying the base model weights.","pith_extraction_headline":"DealMaTe transfers materials across objects using depth, normal, and lighting images in a text-free diffusion framework."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15681/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:33:29.856677Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T19:31:19.128901Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T19:21:29.511010Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:21:56.052858Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"5fde6e694c7a5bda59708b423e04d77c6d653436737b719690fc43023d4a704e"},"references":{"count":71,"sample":[{"doi":"","year":2020,"title":"Louis-Philippe Asselin, Denis Laurendeau, and Jean-Francois Lalonde. 2020. Deep SVBRDF estimation on real materials. InInternational Conference on 3D Vision (3DV). IEEE, 1157–1166","work_id":"6c4a8944-95f4-4da1-8ed5-70ffab6c1c1a","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Tim Brooks, Aleksander Holynski, and Alexei A Efros. 2023. Instructpix2pix: Learning to follow image editing instructions. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog","work_id":"57112cb9-14a4-42c2-9040-91f2a485b532","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"George Cazenavette, Tongzhou Wang, Antonio Torralba, Alexei A Efros, and Jun-Yan Zhu. 2022. Wearable ImageNet: Synthesizing tileable textures via dataset distillation. InProceedings of the IEEE/CVF Co","work_id":"258db9bb-a3d1-4d01-a49b-03a0628cc4d7","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Dave Zhenyu Chen, Yawar Siddiqui, Hsin-Ying Lee, Sergey Tulyakov, and Matthias Nießner. 2023. Text2tex: Text-driven texture synthesis via diffusion models. In Proceedings of the IEEE/CVF International","work_id":"66614e8c-0064-4370-8057-7070127b4e76","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Junsong Chen, YU Jincheng, GE Chongjian, Lewei Yao, Enze Xie, Zhongdao Wang, James Kwok, Ping Luo, Huchuan Lu, and Zhenguo Li. 2024. Pixart-𝛼: Fast training of diffusion transformer for photorealistic","work_id":"0eb81963-2666-40cd-bd12-225119e4f7d6","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":71,"snapshot_sha256":"b4378e7296761f1ac48359a62d72802d91ad313534fa275827f8606ddc160c00","internal_anchors":2},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}