{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:JHZEK4RYR6EIMYH74LAL2UTT2R","short_pith_number":"pith:JHZEK4RY","schema_version":"1.0","canonical_sha256":"49f24572388f888660ffe2c0bd5273d4657a742539259bcecfd0a5214cc94674","source":{"kind":"arxiv","id":"1906.06792","version":1},"attestation_state":"computed","paper":{"title":"Floors are Flat: Leveraging Semantics for Real-Time Surface Normal Prediction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alireza Fathi, Irfan Essa, Karthik Raveendran, Kevin Murphy, Steven Hickson","submitted_at":"2019-06-16T23:01:32Z","abstract_excerpt":"We propose 4 insights that help to significantly improve the performance of deep learning models that predict surface normals and semantic labels from a single RGB image. These insights are: (1) denoise the \"ground truth\" surface normals in the training set to ensure consistency with the semantic labels; (2) concurrently train on a mix of real and synthetic data, instead of pretraining on synthetic and finetuning on real; (3) jointly predict normals and semantics using a shared model, but only backpropagate errors on pixels that have valid training labels; (4) slim down the model and use grays"},"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":"1906.06792","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2019-06-16T23:01:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"22e66b6c4bfac8cba3db501d8b385d6d5e7890e6e602e096d2ff00142ed83ca0","abstract_canon_sha256":"0a8eec8ff1d5ffb4ae79ea57cf998d51014bfff960568f8641782f7082ebe466"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:12.646808Z","signature_b64":"+idtsNZAX6fWVrMg1X0CQuQoq+lx/SAlYsg/dKqO6SLK5ifB0uyti6MKekFk5IusfuhDlhKlDZ8p/hh9Y0hdDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"49f24572388f888660ffe2c0bd5273d4657a742539259bcecfd0a5214cc94674","last_reissued_at":"2026-05-17T23:43:12.646181Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:12.646181Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Floors are Flat: Leveraging Semantics for Real-Time Surface Normal Prediction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alireza Fathi, Irfan Essa, Karthik Raveendran, Kevin Murphy, Steven Hickson","submitted_at":"2019-06-16T23:01:32Z","abstract_excerpt":"We propose 4 insights that help to significantly improve the performance of deep learning models that predict surface normals and semantic labels from a single RGB image. These insights are: (1) denoise the \"ground truth\" surface normals in the training set to ensure consistency with the semantic labels; (2) concurrently train on a mix of real and synthetic data, instead of pretraining on synthetic and finetuning on real; (3) jointly predict normals and semantics using a shared model, but only backpropagate errors on pixels that have valid training labels; (4) slim down the model and use grays"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.06792","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":""},"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":"1906.06792","created_at":"2026-05-17T23:43:12.646274+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.06792v1","created_at":"2026-05-17T23:43:12.646274+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.06792","created_at":"2026-05-17T23:43:12.646274+00:00"},{"alias_kind":"pith_short_12","alias_value":"JHZEK4RYR6EI","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_16","alias_value":"JHZEK4RYR6EIMYH7","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_8","alias_value":"JHZEK4RY","created_at":"2026-05-18T12:33:21.387695+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R","json":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R.json","graph_json":"https://pith.science/api/pith-number/JHZEK4RYR6EIMYH74LAL2UTT2R/graph.json","events_json":"https://pith.science/api/pith-number/JHZEK4RYR6EIMYH74LAL2UTT2R/events.json","paper":"https://pith.science/paper/JHZEK4RY"},"agent_actions":{"view_html":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R","download_json":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R.json","view_paper":"https://pith.science/paper/JHZEK4RY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.06792&json=true","fetch_graph":"https://pith.science/api/pith-number/JHZEK4RYR6EIMYH74LAL2UTT2R/graph.json","fetch_events":"https://pith.science/api/pith-number/JHZEK4RYR6EIMYH74LAL2UTT2R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R/action/storage_attestation","attest_author":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R/action/author_attestation","sign_citation":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R/action/citation_signature","submit_replication":"https://pith.science/pith/JHZEK4RYR6EIMYH74LAL2UTT2R/action/replication_record"}},"created_at":"2026-05-17T23:43:12.646274+00:00","updated_at":"2026-05-17T23:43:12.646274+00:00"}