{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:KYEZSTHYTTUDRW42WYFB4NQM3U","short_pith_number":"pith:KYEZSTHY","schema_version":"1.0","canonical_sha256":"5609994cf89ce838db9ab60a1e360cdd077b49a0869b2a7ea743b0d2ed56deb7","source":{"kind":"arxiv","id":"1702.08502","version":3},"attestation_state":"computed","paper":{"title":"Understanding Convolution for Semantic Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ding Liu, Garrison Cottrell, Panqu Wang, Pengfei Chen, Xiaodi Hou, Ye Yuan, Zehua Huang","submitted_at":"2017-02-27T20:05:11Z","abstract_excerpt":"Recent advances in deep learning, especially deep convolutional neural networks (CNNs), have led to significant improvement over previous semantic segmentation systems. Here we show how to improve pixel-wise semantic segmentation by manipulating convolution-related operations that are of both theoretical and practical value. First, we design dense upsampling convolution (DUC) to generate pixel-level prediction, which is able to capture and decode more detailed information that is generally missing in bilinear upsampling. Second, we propose a hybrid dilated convolution (HDC) framework in the en"},"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":"1702.08502","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-02-27T20:05:11Z","cross_cats_sorted":[],"title_canon_sha256":"0349c2d28ae0ebcd5cf072792f5e6aeb55a23027b2019b359511ab0d4747a489","abstract_canon_sha256":"768c1b014607ee4702b540785d94d6397fb53244e829921acd16b193e104465f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:14:29.137864Z","signature_b64":"sSbW1vLU/Ap3x4JXkwNdpVYmAb3oLXJhYS0UxMzfyUFW4hlfedv0CxqcJL16DDePdMDMBe8MprPialzoy5t8Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5609994cf89ce838db9ab60a1e360cdd077b49a0869b2a7ea743b0d2ed56deb7","last_reissued_at":"2026-05-18T00:14:29.137139Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:14:29.137139Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Convolution for Semantic Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ding Liu, Garrison Cottrell, Panqu Wang, Pengfei Chen, Xiaodi Hou, Ye Yuan, Zehua Huang","submitted_at":"2017-02-27T20:05:11Z","abstract_excerpt":"Recent advances in deep learning, especially deep convolutional neural networks (CNNs), have led to significant improvement over previous semantic segmentation systems. Here we show how to improve pixel-wise semantic segmentation by manipulating convolution-related operations that are of both theoretical and practical value. First, we design dense upsampling convolution (DUC) to generate pixel-level prediction, which is able to capture and decode more detailed information that is generally missing in bilinear upsampling. Second, we propose a hybrid dilated convolution (HDC) framework in the en"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.08502","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":""},"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":"1702.08502","created_at":"2026-05-18T00:14:29.137266+00:00"},{"alias_kind":"arxiv_version","alias_value":"1702.08502v3","created_at":"2026-05-18T00:14:29.137266+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.08502","created_at":"2026-05-18T00:14:29.137266+00:00"},{"alias_kind":"pith_short_12","alias_value":"KYEZSTHYTTUD","created_at":"2026-05-18T12:31:28.150371+00:00"},{"alias_kind":"pith_short_16","alias_value":"KYEZSTHYTTUDRW42","created_at":"2026-05-18T12:31:28.150371+00:00"},{"alias_kind":"pith_short_8","alias_value":"KYEZSTHY","created_at":"2026-05-18T12:31:28.150371+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.03128","citing_title":"Multi-level Wavelet Convolutional Neural Networks","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"1706.05587","citing_title":"Rethinking Atrous Convolution for Semantic Image Segmentation","ref_index":84,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U","json":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U.json","graph_json":"https://pith.science/api/pith-number/KYEZSTHYTTUDRW42WYFB4NQM3U/graph.json","events_json":"https://pith.science/api/pith-number/KYEZSTHYTTUDRW42WYFB4NQM3U/events.json","paper":"https://pith.science/paper/KYEZSTHY"},"agent_actions":{"view_html":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U","download_json":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U.json","view_paper":"https://pith.science/paper/KYEZSTHY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1702.08502&json=true","fetch_graph":"https://pith.science/api/pith-number/KYEZSTHYTTUDRW42WYFB4NQM3U/graph.json","fetch_events":"https://pith.science/api/pith-number/KYEZSTHYTTUDRW42WYFB4NQM3U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U/action/storage_attestation","attest_author":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U/action/author_attestation","sign_citation":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U/action/citation_signature","submit_replication":"https://pith.science/pith/KYEZSTHYTTUDRW42WYFB4NQM3U/action/replication_record"}},"created_at":"2026-05-18T00:14:29.137266+00:00","updated_at":"2026-05-18T00:14:29.137266+00:00"}