pith. sign in
Pith Number

pith:TGYCQ5DM

pith:2026:TGYCQ5DMO4BRN35MNY4RB3RAM3
not attested not anchored not stored refs resolved

M2Retinexformer: Multi-Modal Retinexformer for Low-Light Image Enhancement

Hicham G. Elmongui, Marwan Torki, Youssef Aboelwafa

M2Retinexformer improves low-light image enhancement by fusing depth cues, luminance priors, and semantic features into Retinexformer through multi-scale cross-attention.

arxiv:2605.12556 v1 · 2026-05-11 · cs.CV

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{TGYCQ5DMO4BRN35MNY4RB3RAM3}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Evaluations on the LOL, SID, SMID, and SDSD benchmarks demonstrate overall improvements over Retinexformer and recent state-of-the-art methods.

C2weakest 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.

C3one 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.

References

37 extracted · 37 resolved · 3 Pith anchors

[1] M2Retinexformer: Multi-Modal Retinexformer for Low-Light Image Enhancement 2026 · arXiv:2605.12556
[2] 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
[3] 3, we present the overall architecture of M2Retinexformer, which extends Retinexformer by incor- porating complementary multi-modal cues
[4] 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
[5] 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

Formal links

2 machine-checked theorem links

Cited by

1 paper in Pith

Receipt and verification
First computed 2026-05-18T03:10:02.057752Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

99b028746c770316efac6e3910ee2066f32018b44f1eec9d0239992a27c5dc6a

Aliases

arxiv: 2605.12556 · arxiv_version: 2605.12556v1 · doi: 10.48550/arxiv.2605.12556 · pith_short_12: TGYCQ5DMO4BR · pith_short_16: TGYCQ5DMO4BRN35M · pith_short_8: TGYCQ5DM
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/TGYCQ5DMO4BRN35MNY4RB3RAM3 \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 99b028746c770316efac6e3910ee2066f32018b44f1eec9d0239992a27c5dc6a
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "f178f9fb0ff55b4f1046645dcf32636a6bdf1ddf78d2cc272f8cc9048ad28925",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-11T12:13:13Z",
    "title_canon_sha256": "1b1ba165d5f8eed9576cb4835aec70e3c3ee7550e2cbc541842ee228376ee8eb"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12556",
    "kind": "arxiv",
    "version": 1
  }
}