{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UZWI2T4DLVJSTB2JI6MHPTLTHI","short_pith_number":"pith:UZWI2T4D","schema_version":"1.0","canonical_sha256":"a66c8d4f835d53298749479877cd733a2f309599cbe24b50bc49e49713e0ab76","source":{"kind":"arxiv","id":"2505.00615","version":1},"attestation_state":"computed","paper":{"title":"Pixel3DMM: Versatile Screen-Space Priors for Single-Image 3D Face Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Lourdes Agapito, Martin R\\\"unz, Matthias Nie{\\ss}ner, Simon Giebenhain, Tobias Kirschstein","submitted_at":"2025-05-01T15:47:03Z","abstract_excerpt":"We address the 3D reconstruction of human faces from a single RGB image. To this end, we propose Pixel3DMM, a set of highly-generalized vision transformers which predict per-pixel geometric cues in order to constrain the optimization of a 3D morphable face model (3DMM). We exploit the latent features of the DINO foundation model, and introduce a tailored surface normal and uv-coordinate prediction head. We train our model by registering three high-quality 3D face datasets against the FLAME mesh topology, which results in a total of over 1,000 identities and 976K images. For 3D face reconstruct"},"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":"2505.00615","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-01T15:47:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"91853197e4e687639164922f3d97e853f1a1f21cfacbb040f3cc2e4fca9126a4","abstract_canon_sha256":"d587250a2536dbd253450a536a1f15195366460465e28d73b219937bb462024f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:57:23.039361Z","signature_b64":"1LG85056ZUs1C08u+XmBWWWg56lNKgpd6l+6us4yHuIRVmLeNeALmUC9AhxEc2X8tHj6cRx4ox0uGTLKodr4AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a66c8d4f835d53298749479877cd733a2f309599cbe24b50bc49e49713e0ab76","last_reissued_at":"2026-07-05T10:57:23.038831Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:57:23.038831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pixel3DMM: Versatile Screen-Space Priors for Single-Image 3D Face Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Lourdes Agapito, Martin R\\\"unz, Matthias Nie{\\ss}ner, Simon Giebenhain, Tobias Kirschstein","submitted_at":"2025-05-01T15:47:03Z","abstract_excerpt":"We address the 3D reconstruction of human faces from a single RGB image. To this end, we propose Pixel3DMM, a set of highly-generalized vision transformers which predict per-pixel geometric cues in order to constrain the optimization of a 3D morphable face model (3DMM). We exploit the latent features of the DINO foundation model, and introduce a tailored surface normal and uv-coordinate prediction head. We train our model by registering three high-quality 3D face datasets against the FLAME mesh topology, which results in a total of over 1,000 identities and 976K images. For 3D face reconstruct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00615","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/2505.00615/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":"2505.00615","created_at":"2026-07-05T10:57:23.038890+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.00615v1","created_at":"2026-07-05T10:57:23.038890+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00615","created_at":"2026-07-05T10:57:23.038890+00:00"},{"alias_kind":"pith_short_12","alias_value":"UZWI2T4DLVJS","created_at":"2026-07-05T10:57:23.038890+00:00"},{"alias_kind":"pith_short_16","alias_value":"UZWI2T4DLVJSTB2J","created_at":"2026-07-05T10:57:23.038890+00:00"},{"alias_kind":"pith_short_8","alias_value":"UZWI2T4D","created_at":"2026-07-05T10:57:23.038890+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.19702","citing_title":"Face Anything: 4D Face Reconstruction from Any Image Sequence","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24176","citing_title":"Loki: Representation over Architecture for Diffusion-Based Portrait Animation","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27659","citing_title":"GeoFace: Consistent Multi-View Face Generation with Geometry-Constrained Diffusion","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31283","citing_title":"Topologically Consistent Multi-view 3D Head Reconstruction via Coarse-Guided Layered Surface Sampling","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2511.02830","citing_title":"Densemarks: Learning Canonical Embeddings for Human Heads Images via Point Tracks","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2512.15599","citing_title":"FlexAvatar: Learning Complete 3D Head Avatars with Partial Supervision","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14594","citing_title":"TOPOS: High-Fidelity and Efficient Industry-Grade 3D Head Generation","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06179","citing_title":"SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04524","citing_title":"High-Fidelity Single-Image Head Modeling with Industry-Grade Topology","ref_index":173,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13856","citing_title":"Any3DAvatar: Fast and High-Quality Full-Head 3D Avatar Reconstruction from Single Portrait Image","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19702","citing_title":"Face Anything: 4D Face Reconstruction from Any Image Sequence","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01746","citing_title":"Profile-Specific 3DMM Regression from a Single Lateral Face Image","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI","json":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI.json","graph_json":"https://pith.science/api/pith-number/UZWI2T4DLVJSTB2JI6MHPTLTHI/graph.json","events_json":"https://pith.science/api/pith-number/UZWI2T4DLVJSTB2JI6MHPTLTHI/events.json","paper":"https://pith.science/paper/UZWI2T4D"},"agent_actions":{"view_html":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI","download_json":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI.json","view_paper":"https://pith.science/paper/UZWI2T4D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.00615&json=true","fetch_graph":"https://pith.science/api/pith-number/UZWI2T4DLVJSTB2JI6MHPTLTHI/graph.json","fetch_events":"https://pith.science/api/pith-number/UZWI2T4DLVJSTB2JI6MHPTLTHI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI/action/storage_attestation","attest_author":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI/action/author_attestation","sign_citation":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI/action/citation_signature","submit_replication":"https://pith.science/pith/UZWI2T4DLVJSTB2JI6MHPTLTHI/action/replication_record"}},"created_at":"2026-07-05T10:57:23.038890+00:00","updated_at":"2026-07-05T10:57:23.038890+00:00"}