{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:YO7ZP3KVAACCZD6ENX3MFWSG4K","short_pith_number":"pith:YO7ZP3KV","schema_version":"1.0","canonical_sha256":"c3bf97ed5500042c8fc46df6c2da46e28e99efcb817dec0ff047f8a15208e966","source":{"kind":"arxiv","id":"2405.15757","version":3},"attestation_state":"computed","paper":{"title":"Looking Backward: Streaming Video-to-Video Translation with Feature Banks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MM"],"primary_cat":"cs.CV","authors_text":"Akio Kodaira, Chenfeng Xu, Diana Marculescu, Feng Liang, Kurt Keutzer, Masayoshi Tomizuka","submitted_at":"2024-05-24T17:53:06Z","abstract_excerpt":"This paper introduces StreamV2V, a diffusion model that achieves real-time streaming video-to-video (V2V) translation with user prompts. Unlike prior V2V methods using batches to process limited frames, we opt to process frames in a streaming fashion, to support unlimited frames. At the heart of StreamV2V lies a backward-looking principle that relates the present to the past. This is realized by maintaining a feature bank, which archives information from past frames. For incoming frames, StreamV2V extends self-attention to include banked keys and values and directly fuses similar past 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":"2405.15757","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-24T17:53:06Z","cross_cats_sorted":["cs.MM"],"title_canon_sha256":"6869614bc50caa62bbd797cc15b460f9d73b1ac9eb76fcb91627cf09c80329bf","abstract_canon_sha256":"595037908529a013acc46758c5729f03219778b40d5d199bea7d1536208595e6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:14:40.627570Z","signature_b64":"RxGsnFDxegtHh2cktI6pZ8x1Z/8LZdnCnIaBnSYf/IMxYYywnhUThdRczL64i5jjpaWLam8I1gXjSDnCk7hLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3bf97ed5500042c8fc46df6c2da46e28e99efcb817dec0ff047f8a15208e966","last_reissued_at":"2026-07-05T10:14:40.627072Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:14:40.627072Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Looking Backward: Streaming Video-to-Video Translation with Feature Banks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MM"],"primary_cat":"cs.CV","authors_text":"Akio Kodaira, Chenfeng Xu, Diana Marculescu, Feng Liang, Kurt Keutzer, Masayoshi Tomizuka","submitted_at":"2024-05-24T17:53:06Z","abstract_excerpt":"This paper introduces StreamV2V, a diffusion model that achieves real-time streaming video-to-video (V2V) translation with user prompts. Unlike prior V2V methods using batches to process limited frames, we opt to process frames in a streaming fashion, to support unlimited frames. At the heart of StreamV2V lies a backward-looking principle that relates the present to the past. This is realized by maintaining a feature bank, which archives information from past frames. For incoming frames, StreamV2V extends self-attention to include banked keys and values and directly fuses similar past features"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15757","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2405.15757/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":"2405.15757","created_at":"2026-07-05T10:14:40.627126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.15757v3","created_at":"2026-07-05T10:14:40.627126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15757","created_at":"2026-07-05T10:14:40.627126+00:00"},{"alias_kind":"pith_short_12","alias_value":"YO7ZP3KVAACC","created_at":"2026-07-05T10:14:40.627126+00:00"},{"alias_kind":"pith_short_16","alias_value":"YO7ZP3KVAACCZD6E","created_at":"2026-07-05T10:14:40.627126+00:00"},{"alias_kind":"pith_short_8","alias_value":"YO7ZP3KV","created_at":"2026-07-05T10:14:40.627126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26740","citing_title":"LiveEdit: Towards Real-Time Diffusion-Based Streaming Video Editing","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21466","citing_title":"StreamEdit: Training-Free Video Editing via Few-Step Streaming Video Generation","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21466","citing_title":"StreamEdit: Training-Free Video Editing via Few-Step Streaming Video Generation","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17019","citing_title":"StreamingEffect: Real-Time Human-Centric Video Effect Generation","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19741","citing_title":"CityRAG: Stepping Into a City via Spatially-Grounded Video Generation","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15911","citing_title":"Efficient Video Diffusion Models: Advancements and Challenges","ref_index":76,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K","json":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K.json","graph_json":"https://pith.science/api/pith-number/YO7ZP3KVAACCZD6ENX3MFWSG4K/graph.json","events_json":"https://pith.science/api/pith-number/YO7ZP3KVAACCZD6ENX3MFWSG4K/events.json","paper":"https://pith.science/paper/YO7ZP3KV"},"agent_actions":{"view_html":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K","download_json":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K.json","view_paper":"https://pith.science/paper/YO7ZP3KV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.15757&json=true","fetch_graph":"https://pith.science/api/pith-number/YO7ZP3KVAACCZD6ENX3MFWSG4K/graph.json","fetch_events":"https://pith.science/api/pith-number/YO7ZP3KVAACCZD6ENX3MFWSG4K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K/action/storage_attestation","attest_author":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K/action/author_attestation","sign_citation":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K/action/citation_signature","submit_replication":"https://pith.science/pith/YO7ZP3KVAACCZD6ENX3MFWSG4K/action/replication_record"}},"created_at":"2026-07-05T10:14:40.627126+00:00","updated_at":"2026-07-05T10:14:40.627126+00:00"}