{"paper":{"title":"Make-It-Poseable: Feed-forward Latent Posing Model for 3D Characters","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Make-It-Poseable poses 3D characters by transforming compact latent representations instead of meshes or skinning weights.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alan Zhao, Houqiang Li, Jax Xiang, Ori Zhang, Wengang Zhou, Zhenxun Yuan, Zhiyang Guo","submitted_at":"2025-12-18T17:01:44Z","abstract_excerpt":"Posing 3D characters is a fundamental task in computer graphics. However, existing paradigms, ranging from traditional auto-rigging to recent pose-conditioned generative models, frequently struggle with inaccurate skinning weights, fixed mesh topologies, and poor pose conformance. These challenges have become particularly pronounced with the recent explosion of AI-generated 3D assets, which often exhibit flawed structures and fused geometry. To address these issues, we introduce Make-It-Poseable, a novel feed-forward framework that reformulates character posing as a skinning-free latent-space "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"our method significantly outperforms existing baselines in posing quality. Furthermore, our skeleton-agnostic design exhibits remarkable zero-shot generalization to diverse morphologies including quadrupeds and seamlessly supports various 3D authoring applications such as part replacement and refinement.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That operating directly on compact latent representations can faithfully reconstruct fine geometric details and handle arbitrary topological changes without introducing artifacts or requiring mesh-specific priors.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A latent-space transformer framework poses 3D characters without skinning or fixed topologies, outperforming baselines and generalizing zero-shot to quadrupeds.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Make-It-Poseable poses 3D characters by transforming compact latent representations instead of meshes or skinning weights.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"7df238a737d1a0ceacea91d29529ade62ff2261f29bd1958960570beab072a62"},"source":{"id":"2512.16767","kind":"arxiv","version":2},"verdict":{"id":"87371eba-2b3d-42f8-b4e0-c2acf6ea6f17","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T21:27:44.744095Z","strongest_claim":"our method significantly outperforms existing baselines in posing quality. Furthermore, our skeleton-agnostic design exhibits remarkable zero-shot generalization to diverse morphologies including quadrupeds and seamlessly supports various 3D authoring applications such as part replacement and refinement.","one_line_summary":"A latent-space transformer framework poses 3D characters without skinning or fixed topologies, outperforming baselines and generalizing zero-shot to quadrupeds.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That operating directly on compact latent representations can faithfully reconstruct fine geometric details and handle arbitrary topological changes without introducing artifacts or requiring mesh-specific priors.","pith_extraction_headline":"Make-It-Poseable poses 3D characters by transforming compact latent representations instead of meshes or skinning weights."},"references":{"count":51,"sample":[{"doi":"","year":2024,"title":"Adobe. Mixamo, 2024.https://www.mixamo.com. 6","work_id":"608ccfb9-3ac3-4396-91e8-e4ebdf6797e8","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2007,"title":"Automatic rigging and ani- mation of 3D characters.ACM TOG, 26(3):72–es, 2007","work_id":"bc3709f2-d75c-4703-a066-c3f2cc839449","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Human- Rig: Learning automatic rigging for humanoid character in a large scale dataset, 2024","work_id":"7d6e92a8-8dcb-4d27-ac25-82a8c35b3415","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"DetailGen3D: Generative 3D geometry enhancement via data-dependent flow, 2025","work_id":"2d63e80b-e2df-446f-b443-f1d4ae4b4aea","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Anymate: A dataset and baselines for learning 3D object rigging","work_id":"2f65790f-4040-4cb8-bee1-55ce0de48a7b","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":51,"snapshot_sha256":"53dfa692ee62969968dac592a450a102cc930a066a4c3506242460b52be034a7","internal_anchors":3},"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"}