SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2505.23068 · CVPR · 2025

URWKV: Unified RWKV Model with Multi-state Perspective for Low-light Image Restoration

Rui Xu, Yuzhong Chen, Yuzhen Niu, Huangbiao Xu, Yuezhou Li, Wenxi Liu

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 13 functions out of this paper's own repositories and ran 12 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
FZU-N/URWKV canonical 12 of 13
FunctionStatusWhere it lives
calculate_psnr Ran FZU-N/URWKV/custom_utils/image_utils.py
code served (permissive licence) · get_code("d292dd965ab59629")
check_keys Ran FZU-N/URWKV/custom_utils/model_load.py
code served (permissive licence) · get_code("ec73a18eb6cee64f")
create_window Ran FZU-N/URWKV/model/loss.py
code served (permissive licence) · get_code("6154e3744ece5728")
gaussian Ran FZU-N/URWKV/custom_utils/losses.py
code served (permissive licence) · get_code("b98ab675041aad53")
get_gaussian_kernel Ran FZU-N/URWKV/custom_utils/losses.py
code served (permissive licence) · get_code("7f494e37955dbf6d")
get_gaussian_kernel2d Ran FZU-N/URWKV/custom_utils/losses.py
code served (permissive licence) · get_code("f604cce48cb1aaf6")
is_frozen Ran FZU-N/URWKV/custom_utils/model_utils.py
code served (permissive licence) · get_code("930cddf0a4d80ab1")
load_img Ran FZU-N/URWKV/custom_utils/image_utils.py
code served (permissive licence) · get_code("04f9122f3967b7ae")
load_start_epoch Ran FZU-N/URWKV/custom_utils/model_utils.py
code served (permissive licence) · get_code("c3bea8c4ee6bef4a")
network_parameters Ran FZU-N/URWKV/custom_utils/model_utils.py
code served (permissive licence) · get_code("92e01b46692e513c")
q_shift Ran FZU-N/URWKV/model/modules/urwkv.py
code served (permissive licence) · get_code("a2d0a36f82b566d4")
remove_prefix Ran FZU-N/URWKV/custom_utils/model_load.py
code served (permissive licence) · get_code("a57313630ecbb3a1")
load_pretrain Not yet run FZU-N/URWKV/custom_utils/model_load.py
code served (permissive licence) · get_code("1023e0f5057a4383")

Repositories linked to this paper

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

Abstract

Existing low-light image enhancement (LLIE) and joint LLIE and deblurring (LLIE-deblur) models have made strides in addressing predefined degradations, yet they are often constrained by dynamically coupled degradations. To address these challenges, we introduce a Unified Receptance Weighted Key Value (URWKV) model with multistate perspective, enabling flexible and effective degradation restoration for low-light images. Specifically, we customize the core URWKV block to perceive and analyze complex degradations by leveraging multiple intra-and interstage states. First, inspired by the pupil mechanism in the human visual system, we propose Luminance-adaptive Normalization (LAN) that adjusts normalization parameters based on rich inter-stage states, allowing for adaptive, scene-aware luminance modulation. Second, we aggregate multiple intra-stage states through exponential moving average approach, effectively capturing subtle variations while mitigating information loss inherent in the singlestate mechanism. To reduce the degradation effects commonly associated with conventional skip connections, we propose the State-aware Selective Fusion (SSF) module, which dynamically aligns and integrates multi-state features across encoder stages, selectively fusing contextual information. In comparison to state-of-the-art models, our URWKV model achieves superior performance on various benchmarks, while requiring significantly fewer parameters and computational resources. Code is available at: https://github.com/FZU-N/URWKV.

For agents

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

get_harvested_code_for_paper("2505.23068")
get_code_for_paper("2505.23068")
have("2505.23068")

Connect an agent — have() is free.