{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RHACHDTH6XCMUTQAVCGV5WMDPG","short_pith_number":"pith:RHACHDTH","schema_version":"1.0","canonical_sha256":"89c0238e67f5c4ca4e00a88d5ed983798417e4b56cc63e70d790b2d392bb8308","source":{"kind":"arxiv","id":"2502.03836","version":1},"attestation_state":"computed","paper":{"title":"Adapting Human Mesh Recovery with Vision-Language Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Buzhen Huang, Chengfang Zhang, Chongyang Xu, Yangang Wang, Ziliang Feng","submitted_at":"2025-02-06T07:42:00Z","abstract_excerpt":"Human mesh recovery can be approached using either regression-based or optimization-based methods. Regression models achieve high pose accuracy but struggle with model-to-image alignment due to the lack of explicit 2D-3D correspondences. In contrast, optimization-based methods align 3D models to 2D observations but are prone to local minima and depth ambiguity. In this work, we leverage large vision-language models (VLMs) to generate interactive body part descriptions, which serve as implicit constraints to enhance 3D perception and limit the optimization space. Specifically, we formulate mono"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2502.03836","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-06T07:42:00Z","cross_cats_sorted":[],"title_canon_sha256":"91872fdb1f041c8807ec31b7d8e668ec6385ec5fb0a64bbffc5061d46bd06074","abstract_canon_sha256":"52d00ac19c2969d9430d41839fc9492c633e84db998cabaffbffa437aff54448"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:15.652474Z","signature_b64":"pL/xL2ItL0scg+qr2PHalU2rub4lboIlAiK9wHpSFPfq4P3DCJeBJY1QfQCUEPtqMQt/g4ssyH+Q0wL8nHTSDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89c0238e67f5c4ca4e00a88d5ed983798417e4b56cc63e70d790b2d392bb8308","last_reissued_at":"2026-07-05T10:10:15.652017Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:15.652017Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adapting Human Mesh Recovery with Vision-Language Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Buzhen Huang, Chengfang Zhang, Chongyang Xu, Yangang Wang, Ziliang Feng","submitted_at":"2025-02-06T07:42:00Z","abstract_excerpt":"Human mesh recovery can be approached using either regression-based or optimization-based methods. Regression models achieve high pose accuracy but struggle with model-to-image alignment due to the lack of explicit 2D-3D correspondences. In contrast, optimization-based methods align 3D models to 2D observations but are prone to local minima and depth ambiguity. In this work, we leverage large vision-language models (VLMs) to generate interactive body part descriptions, which serve as implicit constraints to enhance 3D perception and limit the optimization space. Specifically, we formulate mono"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.03836","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2502.03836/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2502.03836","created_at":"2026-07-05T10:10:15.652081+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.03836v1","created_at":"2026-07-05T10:10:15.652081+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.03836","created_at":"2026-07-05T10:10:15.652081+00:00"},{"alias_kind":"pith_short_12","alias_value":"RHACHDTH6XCM","created_at":"2026-07-05T10:10:15.652081+00:00"},{"alias_kind":"pith_short_16","alias_value":"RHACHDTH6XCMUTQA","created_at":"2026-07-05T10:10:15.652081+00:00"},{"alias_kind":"pith_short_8","alias_value":"RHACHDTH","created_at":"2026-07-05T10:10:15.652081+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.13581","citing_title":"SocialMirror: Reconstructing 3D Human Interaction Behaviors from Monocular Videos with Semantic and Geometric Guidance","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG","json":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG.json","graph_json":"https://pith.science/api/pith-number/RHACHDTH6XCMUTQAVCGV5WMDPG/graph.json","events_json":"https://pith.science/api/pith-number/RHACHDTH6XCMUTQAVCGV5WMDPG/events.json","paper":"https://pith.science/paper/RHACHDTH"},"agent_actions":{"view_html":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG","download_json":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG.json","view_paper":"https://pith.science/paper/RHACHDTH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.03836&json=true","fetch_graph":"https://pith.science/api/pith-number/RHACHDTH6XCMUTQAVCGV5WMDPG/graph.json","fetch_events":"https://pith.science/api/pith-number/RHACHDTH6XCMUTQAVCGV5WMDPG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG/action/storage_attestation","attest_author":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG/action/author_attestation","sign_citation":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG/action/citation_signature","submit_replication":"https://pith.science/pith/RHACHDTH6XCMUTQAVCGV5WMDPG/action/replication_record"}},"created_at":"2026-07-05T10:10:15.652081+00:00","updated_at":"2026-07-05T10:10:15.652081+00:00"}