{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:5OXSCIZ5N3IJ3GBRTOGA3OWJRO","short_pith_number":"pith:5OXSCIZ5","schema_version":"1.0","canonical_sha256":"ebaf21233d6ed09d98319b8c0dbac98b97e6655b03b45e063a7e5fe059514bd8","source":{"kind":"arxiv","id":"1604.00133","version":1},"attestation_state":"computed","paper":{"title":"Good Practice in CNN Feature Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jingdong Wang, Liang Zheng, Qi Tian, Shengjin Wang, Yali Zhao","submitted_at":"2016-04-01T05:31:57Z","abstract_excerpt":"The objective of this paper is the effective transfer of the Convolutional Neural Network (CNN) feature in image search and classification. Systematically, we study three facts in CNN transfer. 1) We demonstrate the advantage of using images with a properly large size as input to CNN instead of the conventionally resized one. 2) We benchmark the performance of different CNN layers improved by average/max pooling on the feature maps. Our observation suggests that the Conv5 feature yields very competitive accuracy under such pooling step. 3) We find that the simple combination of pooled features"},"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":"1604.00133","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-04-01T05:31:57Z","cross_cats_sorted":[],"title_canon_sha256":"f97ff991276857bd764b30fab2a568c91b60b4df1a6b92a77245115362c5a972","abstract_canon_sha256":"6bd242fdf0ef4e34a78638dd7f3972637bcd899b82d738069e04d904011304c6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:17:54.470460Z","signature_b64":"LW7FOLyjA1ciO8VrNQI5jKEqR4X8+x+urMzYQJ9htEnby/B+CwDi+vrutHLwa2LkhCRkfQUTE/nuebURmr7sCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ebaf21233d6ed09d98319b8c0dbac98b97e6655b03b45e063a7e5fe059514bd8","last_reissued_at":"2026-05-18T01:17:54.469559Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:17:54.469559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Good Practice in CNN Feature Transfer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jingdong Wang, Liang Zheng, Qi Tian, Shengjin Wang, Yali Zhao","submitted_at":"2016-04-01T05:31:57Z","abstract_excerpt":"The objective of this paper is the effective transfer of the Convolutional Neural Network (CNN) feature in image search and classification. Systematically, we study three facts in CNN transfer. 1) We demonstrate the advantage of using images with a properly large size as input to CNN instead of the conventionally resized one. 2) We benchmark the performance of different CNN layers improved by average/max pooling on the feature maps. Our observation suggests that the Conv5 feature yields very competitive accuracy under such pooling step. 3) We find that the simple combination of pooled features"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1604.00133","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":"1604.00133","created_at":"2026-05-18T01:17:54.469688+00:00"},{"alias_kind":"arxiv_version","alias_value":"1604.00133v1","created_at":"2026-05-18T01:17:54.469688+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1604.00133","created_at":"2026-05-18T01:17:54.469688+00:00"},{"alias_kind":"pith_short_12","alias_value":"5OXSCIZ5N3IJ","created_at":"2026-05-18T12:30:01.593930+00:00"},{"alias_kind":"pith_short_16","alias_value":"5OXSCIZ5N3IJ3GBR","created_at":"2026-05-18T12:30:01.593930+00:00"},{"alias_kind":"pith_short_8","alias_value":"5OXSCIZ5","created_at":"2026-05-18T12:30:01.593930+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.07844","citing_title":"Growing a Brain: Fine-Tuning by Increasing Model Capacity","ref_index":58,"is_internal_anchor":true},{"citing_arxiv_id":"2504.13181","citing_title":"Perception Encoder: The best visual embeddings are not at the output of the network","ref_index":166,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO","json":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO.json","graph_json":"https://pith.science/api/pith-number/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/graph.json","events_json":"https://pith.science/api/pith-number/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/events.json","paper":"https://pith.science/paper/5OXSCIZ5"},"agent_actions":{"view_html":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO","download_json":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO.json","view_paper":"https://pith.science/paper/5OXSCIZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1604.00133&json=true","fetch_graph":"https://pith.science/api/pith-number/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/graph.json","fetch_events":"https://pith.science/api/pith-number/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/action/storage_attestation","attest_author":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/action/author_attestation","sign_citation":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/action/citation_signature","submit_replication":"https://pith.science/pith/5OXSCIZ5N3IJ3GBRTOGA3OWJRO/action/replication_record"}},"created_at":"2026-05-18T01:17:54.469688+00:00","updated_at":"2026-05-18T01:17:54.469688+00:00"}