{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TGYCQ5DMO4BRN35MNY4RB3RAM3","short_pith_number":"pith:TGYCQ5DM","schema_version":"1.0","canonical_sha256":"99b028746c770316efac6e3910ee2066f32018b44f1eec9d0239992a27c5dc6a","source":{"kind":"arxiv","id":"2605.12556","version":1},"attestation_state":"computed","paper":{"title":"M2Retinexformer: Multi-Modal Retinexformer for Low-Light Image Enhancement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"M2Retinexformer improves low-light image enhancement by fusing depth cues, luminance priors, and semantic features into Retinexformer through multi-scale cross-attention.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hicham G. Elmongui, Marwan Torki, Youssef Aboelwafa","submitted_at":"2026-05-11T12:13:13Z","abstract_excerpt":"Low-light image enhancement is challenging due to complex degradations, including amplified noise, artifacts, and color distortion. While Retinex-based deep learning methods have achieved promising results, they primarily rely on single-modality RGB information. We propose M2Retinexformer (Multi-Modal Retinexformer), a novel framework that extends Retinexformer by incorporating depth cues, luminance priors, and semantic features within a progressive refinement pipeline. Depth provides geometric context that is invariant to lighting variations, while luminance and semantic features offer explic"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.12556","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-11T12:13:13Z","cross_cats_sorted":[],"title_canon_sha256":"1b1ba165d5f8eed9576cb4835aec70e3c3ee7550e2cbc541842ee228376ee8eb","abstract_canon_sha256":"f178f9fb0ff55b4f1046645dcf32636a6bdf1ddf78d2cc272f8cc9048ad28925"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:10:02.058409Z","signature_b64":"JoeGcHt9c5nlarRMZ7WPPWcOewhZywvOKoujxF+W3+9DyBoqXmXC7qTRZGXkf1QnQZE52yTM/75E+idnOUwMBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"99b028746c770316efac6e3910ee2066f32018b44f1eec9d0239992a27c5dc6a","last_reissued_at":"2026-05-18T03:10:02.057752Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:10:02.057752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M2Retinexformer: Multi-Modal Retinexformer for Low-Light Image Enhancement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"M2Retinexformer improves low-light image enhancement by fusing depth cues, luminance priors, and semantic features into Retinexformer through multi-scale cross-attention.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hicham G. Elmongui, Marwan Torki, Youssef Aboelwafa","submitted_at":"2026-05-11T12:13:13Z","abstract_excerpt":"Low-light image enhancement is challenging due to complex degradations, including amplified noise, artifacts, and color distortion. While Retinex-based deep learning methods have achieved promising results, they primarily rely on single-modality RGB information. We propose M2Retinexformer (Multi-Modal Retinexformer), a novel framework that extends Retinexformer by incorporating depth cues, luminance priors, and semantic features within a progressive refinement pipeline. Depth provides geometric context that is invariant to lighting variations, while luminance and semantic features offer explic"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Evaluations on the LOL, SID, SMID, and SDSD benchmarks demonstrate overall improvements over Retinexformer and recent state-of-the-art methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That depth cues, luminance priors, and semantic features extracted at multiple scales will remain reliable and provide net positive guidance without introducing new artifacts or requiring perfectly aligned auxiliary data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"M2Retinexformer improves low-light images by progressively refining RGB data with depth, luminance, and semantic modalities through cross-attention and adaptive gating, showing gains on LOL, SID, SMID, and SDSD benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"M2Retinexformer improves low-light image enhancement by fusing depth cues, luminance priors, and semantic features into Retinexformer through multi-scale cross-attention.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"385473aefdb93c434e7e1dc0544b2511a89c9ee5ece77b1c5eb536e58591fc9b"},"source":{"id":"2605.12556","kind":"arxiv","version":1},"verdict":{"id":"0180f31b-e3bb-4a77-bf95-13e456a10824","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:28:28.673488Z","strongest_claim":"Evaluations on the LOL, SID, SMID, and SDSD benchmarks demonstrate overall improvements over Retinexformer and recent state-of-the-art methods.","one_line_summary":"M2Retinexformer improves low-light images by progressively refining RGB data with depth, luminance, and semantic modalities through cross-attention and adaptive gating, showing gains on LOL, SID, SMID, and SDSD benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That depth cues, luminance priors, and semantic features extracted at multiple scales will remain reliable and provide net positive guidance without introducing new artifacts or requiring perfectly aligned auxiliary data.","pith_extraction_headline":"M2Retinexformer improves low-light image enhancement by fusing depth cues, luminance priors, and semantic features into Retinexformer through multi-scale cross-attention."},"references":{"count":37,"sample":[{"doi":"","year":2026,"title":"M2Retinexformer: Multi-Modal Retinexformer for Low-Light Image Enhancement","work_id":"14f8cc97-37c1-410d-b13c-80932d7fa977","ref_index":1,"cited_arxiv_id":"2605.12556","is_internal_anchor":true},{"doi":"","year":null,"title":"Classical ap- proaches such as [8, 9, 10] rely on hand-crafted priors and assume that low-light images are corruption-free, leading to noise amplification and color distortion","work_id":"e0222fdd-7415-4a30-ad41-2e0a29c5da3f","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"3, we present the overall architecture of M2Retinexformer, which extends Retinexformer by incor- porating complementary multi-modal cues","work_id":"3604e6bb-3af0-409a-9fb4-60b44e82442e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"EXPERIMENTS 4.1. Experimental Setup and Implementation Details Datasets.We evaluated M2Retinexformer on seven low- light benchmarks: LOL-v1 [3], LOL-v2 Real/Synthetic [29], SID [30], SMID [31], and SD","work_id":"86f4d1cc-10fd-488b-ac23-ae8070483b7b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Our key insight is that depth provides geometric context that is robust to il- lumination changes, while luminance and semantic features provide content-aware guidance","work_id":"1bfc4a58-7407-4710-adb9-63bf85013bd5","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":37,"snapshot_sha256":"909a9f17aa6856cf87eca8b47aaa3b0a8a85198070261d6f0c299f0292b7e612","internal_anchors":3},"formal_canon":{"evidence_count":2,"snapshot_sha256":"5dc9b458a3c4d73a67592194494f8503f53d9bd1bafe395bea5dc98989813e88"},"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":"2605.12556","created_at":"2026-05-18T03:10:02.057875+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.12556v1","created_at":"2026-05-18T03:10:02.057875+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12556","created_at":"2026-05-18T03:10:02.057875+00:00"},{"alias_kind":"pith_short_12","alias_value":"TGYCQ5DMO4BR","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"TGYCQ5DMO4BRN35M","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"TGYCQ5DM","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.12556","citing_title":"M2Retinexformer: Multi-Modal Retinexformer for Low-Light Image Enhancement","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3","json":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3.json","graph_json":"https://pith.science/api/pith-number/TGYCQ5DMO4BRN35MNY4RB3RAM3/graph.json","events_json":"https://pith.science/api/pith-number/TGYCQ5DMO4BRN35MNY4RB3RAM3/events.json","paper":"https://pith.science/paper/TGYCQ5DM"},"agent_actions":{"view_html":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3","download_json":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3.json","view_paper":"https://pith.science/paper/TGYCQ5DM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.12556&json=true","fetch_graph":"https://pith.science/api/pith-number/TGYCQ5DMO4BRN35MNY4RB3RAM3/graph.json","fetch_events":"https://pith.science/api/pith-number/TGYCQ5DMO4BRN35MNY4RB3RAM3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3/action/storage_attestation","attest_author":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3/action/author_attestation","sign_citation":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3/action/citation_signature","submit_replication":"https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3/action/replication_record"}},"created_at":"2026-05-18T03:10:02.057875+00:00","updated_at":"2026-05-18T03:10:02.057875+00:00"}