{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QZEFOG4OCAFCF427JQVMZ2HPG3","short_pith_number":"pith:QZEFOG4O","schema_version":"1.0","canonical_sha256":"8648571b8e100a22f35f4c2acce8ef36d4711f8e1e2d09dd08698d3e81b0cb3c","source":{"kind":"arxiv","id":"2408.12569","version":3},"attestation_state":"computed","paper":{"title":"Sapiens: Foundation for Human Vision Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Austin James, Julieta Martinez, Peter Selednik, Rawal Khirodkar, Shunsuke Saito, Stuart Anderson, Su Zhaoen, Timur Bagautdinov","submitted_at":"2024-08-22T17:37:27Z","abstract_excerpt":"We present Sapiens, a family of models for four fundamental human-centric vision tasks -- 2D pose estimation, body-part segmentation, depth estimation, and surface normal prediction. Our models natively support 1K high-resolution inference and are extremely easy to adapt for individual tasks by simply fine-tuning models pretrained on over 300 million in-the-wild human images. We observe that, given the same computational budget, self-supervised pretraining on a curated dataset of human images significantly boosts the performance for a diverse set of human-centric tasks. The resulting models ex"},"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":"2408.12569","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-08-22T17:37:27Z","cross_cats_sorted":[],"title_canon_sha256":"e97b14f5979abaf31b09ebb7f74fc84244b79d4b3a37198ac77ec611b3db566c","abstract_canon_sha256":"37b1f75385ffad5c75a8f588d8df50408ef261e54bb7acef7d83e1672fa7b0ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:59:32.904168Z","signature_b64":"E4S7yQsGMYjSH47eHUE3Il3eKT+baAj0Uu3Vgw1vudJKyKGBNggcLn+z5ffEUkJa2rbyUPKdeTOebVJ5S2a1DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8648571b8e100a22f35f4c2acce8ef36d4711f8e1e2d09dd08698d3e81b0cb3c","last_reissued_at":"2026-07-05T08:59:32.903734Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:59:32.903734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sapiens: Foundation for Human Vision Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Austin James, Julieta Martinez, Peter Selednik, Rawal Khirodkar, Shunsuke Saito, Stuart Anderson, Su Zhaoen, Timur Bagautdinov","submitted_at":"2024-08-22T17:37:27Z","abstract_excerpt":"We present Sapiens, a family of models for four fundamental human-centric vision tasks -- 2D pose estimation, body-part segmentation, depth estimation, and surface normal prediction. Our models natively support 1K high-resolution inference and are extremely easy to adapt for individual tasks by simply fine-tuning models pretrained on over 300 million in-the-wild human images. We observe that, given the same computational budget, self-supervised pretraining on a curated dataset of human images significantly boosts the performance for a diverse set of human-centric tasks. The resulting models ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.12569","kind":"arxiv","version":3},"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/2408.12569/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":"2408.12569","created_at":"2026-07-05T08:59:32.903791+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.12569v3","created_at":"2026-07-05T08:59:32.903791+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.12569","created_at":"2026-07-05T08:59:32.903791+00:00"},{"alias_kind":"pith_short_12","alias_value":"QZEFOG4OCAFC","created_at":"2026-07-05T08:59:32.903791+00:00"},{"alias_kind":"pith_short_16","alias_value":"QZEFOG4OCAFCF427","created_at":"2026-07-05T08:59:32.903791+00:00"},{"alias_kind":"pith_short_8","alias_value":"QZEFOG4O","created_at":"2026-07-05T08:59:32.903791+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13655","citing_title":"Flex4DHuman: Flexible Multi-view Video Diffusion for 4D Human Reconstruction","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21766","citing_title":"BodyReLux: Temporally Consistent Full-Body Video Relighting","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16419","citing_title":"Agentic Pipeline for Self-Synchronized Multiview Joint Angle Monitoring in Uncalibrated Environments","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2511.16719","citing_title":"SAM 3: Segment Anything with Concepts","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2512.09646","citing_title":"VHOI: Controllable Video Generation of Human-Object Interactions from Sparse Trajectories via Motion Densification","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03305","citing_title":"HVG-3D: Bridging Real and Simulation Domains for 3D-Conditional Hand-Object Interaction Video Synthesis","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02764","citing_title":"InverseDraping: Recovering Sewing Patterns from 3D Garment Surfaces via BoxMesh Bridging","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22865","citing_title":"MeshLAM: Feed-Forward One-Shot Animatable Textured Mesh Avatar Reconstruction","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20715","citing_title":"GeoRelight: Learning Joint Geometrical Relighting and Reconstruction with Flexible Multi-Modal Diffusion Transformers","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02094","citing_title":"SignMAE: Segmentation-Driven Self-Supervised Learning for Sign Language Recognition","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3","json":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3.json","graph_json":"https://pith.science/api/pith-number/QZEFOG4OCAFCF427JQVMZ2HPG3/graph.json","events_json":"https://pith.science/api/pith-number/QZEFOG4OCAFCF427JQVMZ2HPG3/events.json","paper":"https://pith.science/paper/QZEFOG4O"},"agent_actions":{"view_html":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3","download_json":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3.json","view_paper":"https://pith.science/paper/QZEFOG4O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.12569&json=true","fetch_graph":"https://pith.science/api/pith-number/QZEFOG4OCAFCF427JQVMZ2HPG3/graph.json","fetch_events":"https://pith.science/api/pith-number/QZEFOG4OCAFCF427JQVMZ2HPG3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3/action/storage_attestation","attest_author":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3/action/author_attestation","sign_citation":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3/action/citation_signature","submit_replication":"https://pith.science/pith/QZEFOG4OCAFCF427JQVMZ2HPG3/action/replication_record"}},"created_at":"2026-07-05T08:59:32.903791+00:00","updated_at":"2026-07-05T08:59:32.903791+00:00"}