{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:XVKM2EXXG5MPWKMSXWVT2VUJ5H","short_pith_number":"pith:XVKM2EXX","schema_version":"1.0","canonical_sha256":"bd54cd12f73758fb2992bdab3d5689e9c20edcbb4f96af5f3b9ce9ff0a0f28e7","source":{"kind":"arxiv","id":"1610.00291","version":2},"attestation_state":"computed","paper":{"title":"Deep Feature Consistent Variational Autoencoder","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guoping Qiu, Ke Sun, Linlin Shen, Xianxu Hou","submitted_at":"2016-10-02T15:48:36Z","abstract_excerpt":"We present a novel method for constructing Variational Autoencoder (VAE). Instead of using pixel-by-pixel loss, we enforce deep feature consistency between the input and the output of a VAE, which ensures the VAE's output to preserve the spatial correlation characteristics of the input, thus leading the output to have a more natural visual appearance and better perceptual quality. Based on recent deep learning works such as style transfer, we employ a pre-trained deep convolutional neural network (CNN) and use its hidden features to define a feature perceptual loss for VAE training. Evaluated "},"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":"1610.00291","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-10-02T15:48:36Z","cross_cats_sorted":[],"title_canon_sha256":"7e6a7029dc7bafbaf20a3a50854b6a62c85fae0e36d2a77ffe6b3664480f369d","abstract_canon_sha256":"9f2d34390b5b64db5ee142300da4a1f85ee8d5c8c06ff18a5f7c05a6ea6c4e82"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:58:15.710633Z","signature_b64":"kf2NUiPUsWUqzUpkBdf+TvV+/7CvNxWkx+BqStt+kUe63fRGMQOBI5/JEfpB3tkzL8xHxXn/ZOrpqeaW0ElYCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd54cd12f73758fb2992bdab3d5689e9c20edcbb4f96af5f3b9ce9ff0a0f28e7","last_reissued_at":"2026-07-05T07:58:15.709855Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:58:15.709855Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Feature Consistent Variational Autoencoder","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guoping Qiu, Ke Sun, Linlin Shen, Xianxu Hou","submitted_at":"2016-10-02T15:48:36Z","abstract_excerpt":"We present a novel method for constructing Variational Autoencoder (VAE). Instead of using pixel-by-pixel loss, we enforce deep feature consistency between the input and the output of a VAE, which ensures the VAE's output to preserve the spatial correlation characteristics of the input, thus leading the output to have a more natural visual appearance and better perceptual quality. Based on recent deep learning works such as style transfer, we employ a pre-trained deep convolutional neural network (CNN) and use its hidden features to define a feature perceptual loss for VAE training. Evaluated "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1610.00291","kind":"arxiv","version":2},"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/1610.00291/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":"1610.00291","created_at":"2026-07-05T07:58:15.709925+00:00"},{"alias_kind":"arxiv_version","alias_value":"1610.00291v2","created_at":"2026-07-05T07:58:15.709925+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1610.00291","created_at":"2026-07-05T07:58:15.709925+00:00"},{"alias_kind":"pith_short_12","alias_value":"XVKM2EXXG5MP","created_at":"2026-07-05T07:58:15.709925+00:00"},{"alias_kind":"pith_short_16","alias_value":"XVKM2EXXG5MPWKMS","created_at":"2026-07-05T07:58:15.709925+00:00"},{"alias_kind":"pith_short_8","alias_value":"XVKM2EXX","created_at":"2026-07-05T07:58:15.709925+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.20195","citing_title":"Photometric Super-Resolution for Improving Galaxy Morphological Measurements using Conditional Generative Adversarial Networks","ref_index":140,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H","json":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H.json","graph_json":"https://pith.science/api/pith-number/XVKM2EXXG5MPWKMSXWVT2VUJ5H/graph.json","events_json":"https://pith.science/api/pith-number/XVKM2EXXG5MPWKMSXWVT2VUJ5H/events.json","paper":"https://pith.science/paper/XVKM2EXX"},"agent_actions":{"view_html":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H","download_json":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H.json","view_paper":"https://pith.science/paper/XVKM2EXX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1610.00291&json=true","fetch_graph":"https://pith.science/api/pith-number/XVKM2EXXG5MPWKMSXWVT2VUJ5H/graph.json","fetch_events":"https://pith.science/api/pith-number/XVKM2EXXG5MPWKMSXWVT2VUJ5H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H/action/storage_attestation","attest_author":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H/action/author_attestation","sign_citation":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H/action/citation_signature","submit_replication":"https://pith.science/pith/XVKM2EXXG5MPWKMSXWVT2VUJ5H/action/replication_record"}},"created_at":"2026-07-05T07:58:15.709925+00:00","updated_at":"2026-07-05T07:58:15.709925+00:00"}