PASDiff: Physics-Aware Semantic Guidance for Joint Real-world Low-Light Face Enhancement and Restoration

• URL: http://arxiv.org/abs/2603.24969v1
• Date: Thu, 26 Mar 2026 03:02:39 GMT
• Title: PASDiff: Physics-Aware Semantic Guidance for Joint Real-world Low-Light Face Enhancement and Restoration
• Abstract summary: We propose PASDiff, a Physics-Aware Semantic Diffusion with a training-free manner.<n>We leverage inverse intensity weighting and Retinex theory to introduce photometric constraints, thereby reliably recovering visibility and natural chromaticity.<n>We construct WildDark-Face, a real-world benchmark of 700 low-light facial images with complex degradations.
• Score: 28.276779197981615
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Face images captured in real-world low light suffer multiple degradations-low illumination, blur, noise, and low visibility, etc. Existing cascaded solutions often suffer from severe error accumulation, while generic joint models lack explicit facial priors and struggle to resolve clear face structures. In this paper, we propose PASDiff, a Physics-Aware Semantic Diffusion with a training-free manner. To achieve a plausible illumination and color distribution, we leverage inverse intensity weighting and Retinex theory to introduce photometric constraints, thereby reliably recovering visibility and natural chromaticity. To faithfully reconstruct facial details, our Style-Agnostic Structural Injection (SASI) extracts structures from an off-the-shelf facial prior while filtering out its intrinsic photometric biases, seamlessly harmonizing identity features with physical constraints. Furthermore, we construct WildDark-Face, a real-world benchmark of 700 low-light facial images with complex degradations. Extensive experiments demonstrate that PASDiff significantly outperforms existing methods, achieving a superior balance among natural illumination, color recovery, and identity consistency.

Related papers

• ExpoCM: Exposure-Aware One-Step Generative Single-Image HDR Reconstruction [60.873186669266836]
  Single-image HDR reconstruction aims to recover high dynamic range radiance from a single low dynamic range (LDR) input, but remains highly ill-posed due to detail saturation in over-exposed regions and noise amplification in under-exposed areas.<n>We propose ExpoCM, a novel one-step generative HDR reconstruction framework that reformulates HDR reconstruction as a Probability Flow ODE (PF-ODE) and constructs exposure-aware consistency trajectories via exposure-dependent perturbations.<n>Experiments on the HDR-REAL, HDR-EYE, and AIM2025 benchmarks demonstrate that ExpoCM achieves state-of-the-art fidelity
  arXiv  Detail & Related papers  (2026-05-04T11:06:49Z)
• Exploring Specular Reflection Inconsistency for Generalizable Face Forgery Detection [25.84334069614374]
  forgery detection approaches relying on spatial and frequency features demonstrate limited efficacy against high-quality, entirely synthesized forgeries.<n>We propose a novel detection method grounded in the observation that facial attributes governed by complex physical laws are inherently difficult to replicate.<n>We introduce a fast and accurate face texture estimation method based on Retinex theory to enable precise specular reflection separation.
  arXiv  Detail & Related papers  (2026-02-06T07:27:19Z)
• Near-Light Color Photometric Stereo for mono-Chromaticity non-lambertian surface [67.4383975650003]
  We propose a framework that leverages neural implicit representations for depth and BRDF modeling under the assumption of mono-chromaticity.<n> Experiments on both synthetic and real-world datasets demonstrate that our method achieves accurate and robust surface reconstruction.
  arXiv  Detail & Related papers  (2026-01-19T02:26:08Z)
• Self-Supervised Selective-Guided Diffusion Model for Old-Photo Face Restoration [43.12011252251526]
  Old-photo face restoration poses significant challenges due to compounded degradations such as breakage, fading, and severe blur.<n>We propose Self-Supervised Selective-Guided Diffusion, which leverages pseudo-reference faces generated by a pre-trained diffusion model under weak guidance.<n>SSDiff outperforms existing GAN-based and diffusion-based methods in perceptual quality, fidelity, and regional controllability.
  arXiv  Detail & Related papers  (2025-10-14T03:34:15Z)
• Ordinality of Visible-Thermal Image Intensities for Intrinsic Image Decomposition [25.579052761222513]
  We introduce a training-free approach for intrinsic image decomposition using only a pair of visible and thermal images.<n>We leverage the principle that light not reflected from an opaque surface is absorbed and detected as heat by a thermal camera.<n>This allows us to relate the ordinalities between visible and thermal image intensities to the ordinalities of shading and reflectance.
  arXiv  Detail & Related papers  (2025-09-12T16:29:02Z)
• Degradation-Consistent Learning via Bidirectional Diffusion for Low-Light Image Enhancement [15.619960776528362]
  Low-light image enhancement aims to improve the visibility of degraded images to better align with human visual perception.<n>We propose a bidirectional diffusion optimization mechanism that jointly models the degradation processes of both low-light and normal-light images.<n>Our method outperforms state-of-the-art methods in both quantitative and qualitative evaluations.
  arXiv  Detail & Related papers  (2025-07-24T07:21:23Z)
• Advancing Unsupervised Low-light Image Enhancement: Noise Estimation, Illumination Interpolation, and Self-Regulation [55.07472635587852]
  Low-Light Image Enhancement (LLIE) techniques have made notable advancements in preserving image details and enhancing contrast. These approaches encounter persistent challenges in efficiently mitigating dynamic noise and accommodating diverse low-light scenarios. We first propose a method for estimating the noise level in low light images in a quick and accurate way. We then devise a Learnable Illumination Interpolator (LII) to satisfy general constraints between illumination and input.
  arXiv  Detail & Related papers  (2023-05-17T13:56:48Z)
• LTT-GAN: Looking Through Turbulence by Inverting GANs [86.25869403782957]
  We propose the first turbulence mitigation method that makes use of visual priors encapsulated by a well-trained GAN. Based on the visual priors, we propose to learn to preserve the identity of restored images on a periodic contextual distance. Our method significantly outperforms prior art in both the visual quality and face verification accuracy of restored results.
  arXiv  Detail & Related papers  (2021-12-04T16:42:13Z)
• Towards High Fidelity Monocular Face Reconstruction with Rich Reflectance using Self-supervised Learning and Ray Tracing [49.759478460828504]
  Methods combining deep neural network encoders with differentiable rendering have opened up the path for very fast monocular reconstruction of geometry, lighting and reflectance. ray tracing was introduced for monocular face reconstruction within a classic optimization-based framework. We propose a new method that greatly improves reconstruction quality and robustness in general scenes.
  arXiv  Detail & Related papers  (2021-03-29T08:58:10Z)
• Learning Flow-based Feature Warping for Face Frontalization with Illumination Inconsistent Supervision [73.18554605744842]
  Flow-based Feature Warping Model (FFWM) learns to synthesize photo-realistic and illumination preserving frontal images. An Illumination Preserving Module (IPM) is proposed to learn illumination preserving image synthesis. A Warp Attention Module (WAM) is introduced to reduce the pose discrepancy in the feature level.
  arXiv  Detail & Related papers  (2020-08-16T06:07:00Z)
• Recurrent Exposure Generation for Low-Light Face Detection [113.25331155337759]
  We propose a novel Recurrent Exposure Generation (REG) module and a Multi-Exposure Detection (MED) module. REG produces progressively and efficiently intermediate images corresponding to various exposure settings. Such pseudo-exposures are then fused by MED to detect faces across different lighting conditions.
  arXiv  Detail & Related papers  (2020-07-21T17:30:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.