Learning Reference-Guided Exposure Correction with Hybrid Illumination Characteristics
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910258589335552 |
|---|---|
| author | Ren, Hao Bi, Zetong Wan, Zhaoliang Cheng, Hui |
| author_facet | Ren, Hao Bi, Zetong Wan, Zhaoliang Cheng, Hui |
| contents | We present HICNet, a reference-guided exposure correction framework. A lightweight, content-agnostic encoder distills each image into a compact illumination embedding capturing regional brightness, edge contrast, and higher-order luminance moments. The embedding difference between a source and its reference drives a multi-scale modulation network that combines FiLM-based global adjustment with Photometric Channel Rebalancing for fine-grained, illumination-aware spectral gating, producing exposure-matched outputs while faithfully preserving scene details. A cross-batch contrastive loss orders the illumination manifold, bolstering robustness to diverse lighting conditions. Trained without ground truth or intrinsic decomposition, HICNet attains better accuracy on public benchmarks and generalizes well to entirely unseen scenes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_26729 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning Reference-Guided Exposure Correction with Hybrid Illumination Characteristics Ren, Hao Bi, Zetong Wan, Zhaoliang Cheng, Hui Computer Vision and Pattern Recognition We present HICNet, a reference-guided exposure correction framework. A lightweight, content-agnostic encoder distills each image into a compact illumination embedding capturing regional brightness, edge contrast, and higher-order luminance moments. The embedding difference between a source and its reference drives a multi-scale modulation network that combines FiLM-based global adjustment with Photometric Channel Rebalancing for fine-grained, illumination-aware spectral gating, producing exposure-matched outputs while faithfully preserving scene details. A cross-batch contrastive loss orders the illumination manifold, bolstering robustness to diverse lighting conditions. Trained without ground truth or intrinsic decomposition, HICNet attains better accuracy on public benchmarks and generalizes well to entirely unseen scenes. |
| title | Learning Reference-Guided Exposure Correction with Hybrid Illumination Characteristics |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.26729 |