{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:3HWGLWGOG723HYWDJVB63GLV3L","short_pith_number":"pith:3HWGLWGO","schema_version":"1.0","canonical_sha256":"d9ec65d8ce37f5b3e2c34d43ed9975dad97aa03b10e1eeadb679f6e9d368f03e","source":{"kind":"arxiv","id":"2207.05025","version":1},"attestation_state":"computed","paper":{"title":"PSP-HDRI$+$: A Synthetic Dataset Generator for Pre-Training of Human-Centric Computer Vision Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DB","cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Adam Crespi, Alex Thaman, Chunpu Wang, Maciek Chociej, Salehe Erfanian Ebadi, Sanjay Vishwakarma, Saurav Dhakad, Sujoy Ganguly, You-Cyuan Jhang","submitted_at":"2022-07-11T17:22:58Z","abstract_excerpt":"We introduce a new synthetic data generator PSP-HDRI$+$ that proves to be a superior pre-training alternative to ImageNet and other large-scale synthetic data counterparts. We demonstrate that pre-training with our synthetic data will yield a more general model that performs better than alternatives even when tested on out-of-distribution (OOD) sets. Furthermore, using ablation studies guided by person keypoint estimation metrics with an off-the-shelf model architecture, we show how to manipulate our synthetic data generator to further improve model performance."},"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":"2207.05025","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-07-11T17:22:58Z","cross_cats_sorted":["cs.AI","cs.DB","cs.GR","cs.LG"],"title_canon_sha256":"5885e107cca63df9bda8c1073ef4b0f5ca7819715d2141ae0243febc3f215c8f","abstract_canon_sha256":"e5fc522cc2dcad25d8a00901459f7ce7868d36454915a6a5103568d16ecfcc02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:39:06.097904Z","signature_b64":"ehcfkmfU2WwRFdg/HxTqWWUW8ZhdhiQ1VJ2DWQGgZG3sgtnIAX86PzrjYzz+7ihwQ+psrxbbhwdEB9O6MZQwDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9ec65d8ce37f5b3e2c34d43ed9975dad97aa03b10e1eeadb679f6e9d368f03e","last_reissued_at":"2026-07-05T04:39:06.097439Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:39:06.097439Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PSP-HDRI$+$: A Synthetic Dataset Generator for Pre-Training of Human-Centric Computer Vision Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DB","cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Adam Crespi, Alex Thaman, Chunpu Wang, Maciek Chociej, Salehe Erfanian Ebadi, Sanjay Vishwakarma, Saurav Dhakad, Sujoy Ganguly, You-Cyuan Jhang","submitted_at":"2022-07-11T17:22:58Z","abstract_excerpt":"We introduce a new synthetic data generator PSP-HDRI$+$ that proves to be a superior pre-training alternative to ImageNet and other large-scale synthetic data counterparts. We demonstrate that pre-training with our synthetic data will yield a more general model that performs better than alternatives even when tested on out-of-distribution (OOD) sets. Furthermore, using ablation studies guided by person keypoint estimation metrics with an off-the-shelf model architecture, we show how to manipulate our synthetic data generator to further improve model performance."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.05025","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/2207.05025/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":"2207.05025","created_at":"2026-07-05T04:39:06.097501+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.05025v1","created_at":"2026-07-05T04:39:06.097501+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.05025","created_at":"2026-07-05T04:39:06.097501+00:00"},{"alias_kind":"pith_short_12","alias_value":"3HWGLWGOG723","created_at":"2026-07-05T04:39:06.097501+00:00"},{"alias_kind":"pith_short_16","alias_value":"3HWGLWGOG723HYWD","created_at":"2026-07-05T04:39:06.097501+00:00"},{"alias_kind":"pith_short_8","alias_value":"3HWGLWGO","created_at":"2026-07-05T04:39:06.097501+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.00440","citing_title":"Poplar: A Scalable Pipeline for Human-Centric Image Dataset Synthesis","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L","json":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L.json","graph_json":"https://pith.science/api/pith-number/3HWGLWGOG723HYWDJVB63GLV3L/graph.json","events_json":"https://pith.science/api/pith-number/3HWGLWGOG723HYWDJVB63GLV3L/events.json","paper":"https://pith.science/paper/3HWGLWGO"},"agent_actions":{"view_html":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L","download_json":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L.json","view_paper":"https://pith.science/paper/3HWGLWGO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.05025&json=true","fetch_graph":"https://pith.science/api/pith-number/3HWGLWGOG723HYWDJVB63GLV3L/graph.json","fetch_events":"https://pith.science/api/pith-number/3HWGLWGOG723HYWDJVB63GLV3L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L/action/storage_attestation","attest_author":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L/action/author_attestation","sign_citation":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L/action/citation_signature","submit_replication":"https://pith.science/pith/3HWGLWGOG723HYWDJVB63GLV3L/action/replication_record"}},"created_at":"2026-07-05T04:39:06.097501+00:00","updated_at":"2026-07-05T04:39:06.097501+00:00"}