SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.03349 · 2024

Retinexmamba: Retinex-based Mamba for Low-light Image Enhancement

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 21 functions out of this paper's own repositories and ran 19 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
YhuoyuH/RetinexMamba canonical 19 of 21
FunctionStatusWhere it lives
calc_psnr Ran YhuoyuH/RetinexMamba/Enhancement/RMSE.py
code served (permissive licence) · get_code("f73eaea4d5f18212")
calc_rmse Ran YhuoyuH/RetinexMamba/Enhancement/RMSE.py
code served (permissive licence) · get_code("9fd9f7965e5e9672")
calc_ssim Ran YhuoyuH/RetinexMamba/Enhancement/RMSE.py
code served (permissive licence) · get_code("eb63f9501a5d6798")
calculate_fid Ran YhuoyuH/RetinexMamba/basicsr/metrics/fid.py
code served (permissive licence) · get_code("1d7ce6d6dd7ff5c1")
calculate_psnr Ran YhuoyuH/RetinexMamba/Enhancement/utils.py
code served (permissive licence) · get_code("d292dd965ab59629")
calculate_ssim Ran YhuoyuH/RetinexMamba/Enhancement/utils.py
code served (permissive licence) · get_code("a05b184dc6c264a0")
compute_feature Ran YhuoyuH/RetinexMamba/basicsr/metrics/niqe.py
code served (permissive licence) · get_code("5170ff1c4106b27a")
conv Ran YhuoyuH/RetinexMamba/basicsr/models/archs/RetinexMamba_arch.py
code served (permissive licence) · get_code("4cf5ae1af7c80889")
estimate_aggd_param Ran YhuoyuH/RetinexMamba/basicsr/metrics/niqe.py
code served (permissive licence) · get_code("a1282a48b0941f5d")
extract_inception_features Ran YhuoyuH/RetinexMamba/basicsr/metrics/fid.py
code served (permissive licence) · get_code("c9fb213c957dc82d")
flops_selective_scan_ref Ran YhuoyuH/RetinexMamba/basicsr/models/archs/SS2D_arch.py
code served (permissive licence) · get_code("91788abc3267bdda")
get_position_from_periods Ran YhuoyuH/RetinexMamba/basicsr/models/lr_scheduler.py
code served (permissive licence) · get_code("cd569444547de84f")
prepare_for_ssim Ran YhuoyuH/RetinexMamba/basicsr/metrics/psnr_ssim.py
code served (permissive licence) · get_code("95c6eba9350a455a")
prepare_for_ssim_rgb Ran YhuoyuH/RetinexMamba/basicsr/metrics/psnr_ssim.py
code served (permissive licence) · get_code("fb93712db7064ac4")
reorder_image Ran YhuoyuH/RetinexMamba/basicsr/metrics/metric_util.py
code served (permissive licence) · get_code("95067518dc16b3e5")
shift_back Ran YhuoyuH/RetinexMamba/basicsr/models/archs/RetinexMamba_arch.py
code served (permissive licence) · get_code("7b904d88704f6a90")
to_3d Ran YhuoyuH/RetinexMamba/basicsr/models/archs/IFA_arch.py
code served (permissive licence) · get_code("82a15cc1e46f7e4d")
to_4d Ran YhuoyuH/RetinexMamba/basicsr/models/archs/IFA_arch.py
code served (permissive licence) · get_code("b20f2a5df739a59e")
trunc_normal_ Ran YhuoyuH/RetinexMamba/basicsr/models/archs/RetinexMamba_arch.py
code served (permissive licence) · get_code("afa459e9e9c02c4f")
PSNR Not yet run YhuoyuH/RetinexMamba/Enhancement/utils.py
code served (permissive licence) · get_code("8eccacd101833667")
niqe Not yet run YhuoyuH/RetinexMamba/basicsr/metrics/niqe.py
code served (permissive licence) · get_code("0029cb1566e47350")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In the field of low-light image enhancement, both traditional Retinex methods and advanced deep learning techniques such as Retinexformer have shown distinct advantages and limitations. Traditional Retinex methods, designed to mimic the human eye's perception of brightness and color, decompose images into illumination and reflection components but struggle with noise management and detail preservation under low light conditions. Retinexformer enhances illumination estimation through traditional self-attention mechanisms, but faces challenges with insufficient interpretability and suboptimal enhancement effects. To overcome these limitations, this paper introduces the RetinexMamba architecture. RetinexMamba not only captures the physical intuitiveness of traditional Retinex methods but also integrates the deep learning framework of Retinexformer, leveraging the computational efficiency of State Space Models (SSMs) to enhance processing speed. This architecture features innovative illumination estimators and damage restorer mechanisms that maintain image quality during enhancement. Moreover, RetinexMamba replaces the IG-MSA (Illumination-Guided Multi-Head Attention) in Retinexformer with a Fused-Attention mechanism, improving the model's interpretability. Experimental evaluations on the LOL dataset show that RetinexMamba outperforms existing deep learning approaches based on Retinex theory in both quantitative and qualitative metrics, confirming its effectiveness and superiority in enhancing low-light images.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2405.03349")
get_code_for_paper("2405.03349")
have("2405.03349")

Connect an agent — have() is free.