SIMI is an unsupervised low-light image enhancement network using bit-plane decomposition to mine self-information, reported to reach state-of-the-art performance on standard benchmarks.
Self- referencedeepadaptivecurveestimationforlow-lightimageenhance- ment
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
fields
cs.CV 4verdicts
UNVERDICTED 4representative citing papers
Introduces LSTR and ESTR low-light text datasets and shows joint LLIE-OCR training outperforms standalone models.
LiteIE proposes a two-layer backbone-agnostic feature extractor and parameter-free Iterative Restoration Module for unsupervised low-light enhancement, claiming 19.04 dB PSNR on LOL with 0.07% of SOTA parameters and 30 FPS 4K on Snapdragon 8 Gen 3.
Self-DACE++ enhances low-light images more effectively than prior methods via efficient adaptive adjustment curves, randomized-order training with network fusion, and a Retinex-grounded denoising module while achieving real-time speed.
citing papers explorer
-
SIMI: Self-information Mining Network for Low-light Image Enhancement
SIMI is an unsupervised low-light image enhancement network using bit-plane decomposition to mine self-information, reported to reach state-of-the-art performance on standard benchmarks.
-
Reading in the Dark: Low-light Scene Text Recognition
Introduces LSTR and ESTR low-light text datasets and shows joint LLIE-OCR training outperforms standalone models.
-
Towards Lightest Low-Light Image Enhancement Architecture for Mobile Devices
LiteIE proposes a two-layer backbone-agnostic feature extractor and parameter-free Iterative Restoration Module for unsupervised low-light enhancement, claiming 19.04 dB PSNR on LOL with 0.07% of SOTA parameters and 30 FPS 4K on Snapdragon 8 Gen 3.
-
Self-DACE++: Robust Low-Light Enhancement via Efficient Adaptive Curve Estimation
Self-DACE++ enhances low-light images more effectively than prior methods via efficient adaptive adjustment curves, randomized-order training with network fusion, and a Retinex-grounded denoising module while achieving real-time speed.