Diffusion-Guided Knowledge Distillation for Weakly-Supervised Low-Light Semantic Segmentation

• URL: http://arxiv.org/abs/2507.07578v2
• Date: Wed, 23 Jul 2025 07:51:06 GMT
• Title: Diffusion-Guided Knowledge Distillation for Weakly-Supervised Low-Light Semantic Segmentation
• Authors: Chunyan Wang, Dong Zhang, Jinhui Tang,
• Abstract summary: Weakly-supervised semantic segmentation aims to assign category labels to each pixel using weak annotations.<n>Existing methods significantly degrade in low-light environments due to severe image quality degradation and inherent constraints of weak supervision.<n>We propose Diffusion-Guided Knowledge Distillation for WeaklySupervised Low-light Semantic (DGKD-WLSS)
• Score: 32.4542162399975
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Weakly-supervised semantic segmentation aims to assign category labels to each pixel using weak annotations, significantly reducing manual annotation costs. Although existing methods have achieved remarkable progress in well-lit scenarios, their performance significantly degrades in low-light environments due to two fundamental limitations: severe image quality degradation (e.g., low contrast, noise, and color distortion) and the inherent constraints of weak supervision. These factors collectively lead to unreliable class activation maps and semantically ambiguous pseudo-labels, ultimately compromising the model's ability to learn discriminative feature representations. To address these problems, we propose Diffusion-Guided Knowledge Distillation for Weakly-Supervised Low-light Semantic Segmentation (DGKD-WLSS), a novel framework that synergistically combines Diffusion-Guided Knowledge Distillation (DGKD) with Depth-Guided Feature Fusion (DGF2). DGKD aligns normal-light and low-light features via diffusion-based denoising and knowledge distillation, while DGF2 integrates depth maps as illumination-invariant geometric priors to enhance structural feature learning. Extensive experiments demonstrate the effectiveness of DGKD-WLSS, which achieves state-of-the-art performance in weakly supervised semantic segmentation tasks under low-light conditions. The source codes have been released at:https://github.com/ChunyanWang1/DGKD-WLSS.

Related papers

• 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)
• From Enhancement to Understanding: Build a Generalized Bridge for Low-light Vision via Semantically Consistent Unsupervised Fine-tuning [65.94580484237737]
  Low-light enhancement improves image quality for downstream tasks, but existing methods rely on physical or geometric priors.<n>We build a generalized bridge between low-light enhancement and low-light understanding, which we term Generalized Enhancement For Understanding (GEFU)<n>To address the diverse causes of low-light degradation, we leverage pretrained generative diffusion models to optimize images, achieving zero-shot generalization performance.
  arXiv  Detail & Related papers  (2025-07-11T07:51:26Z)
• Low-Light Enhancement via Encoder-Decoder Network with Illumination Guidance [0.0]
  This paper introduces a novel deep learning framework for low-light image enhancement, named the.<n>the-Decoder Network with Illumination Guidance (EDNIG)<n>EDNIG integrates an illumination map, derived from Bright Channel Prior (BCP), as a guidance input.<n>It is optimized within a Generative Adversarial Network (GAN) framework using a composite loss function that combines adversarial loss, pixel-wise mean squared error (MSE), and perceptual loss.
  arXiv  Detail & Related papers  (2025-07-04T09:35:00Z)
• DSAGL: Dual-Stream Attention-Guided Learning for Weakly Supervised Whole Slide Image Classification [5.260725801393189]
  Whole-slide images (WSIs) are critical for cancer diagnosis due to their ultra-high resolution and rich semantic content.<n>We propose DSAGL (Dual-Stream Attention-Guided Learning), a novel weakly supervised classification framework that combines a teacher-student architecture with a dual-stream design.
  arXiv  Detail & Related papers  (2025-05-29T11:07:16Z)
• Towards Anomaly-Aware Pre-Training and Fine-Tuning for Graph Anomaly Detection [59.042018542376596]
  Graph anomaly detection (GAD) has garnered increasing attention in recent years, yet remains challenging due to two key factors.<n>Anomaly-Aware Pre-Training and Fine-Tuning (APF) is a framework to mitigate the challenges in GAD.<n> Comprehensive experiments on 10 benchmark datasets validate the superior performance of APF in comparison to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-04-19T09:57:35Z)
• Empowering Sparse-Input Neural Radiance Fields with Dual-Level Semantic Guidance from Dense Novel Views [66.1245505423179]
  We show that rendered semantics can be treated as a more robust form of augmented data than rendered RGB.<n>Our method enhances NeRF's performance by incorporating guidance derived from the rendered semantics.
  arXiv  Detail & Related papers  (2025-03-04T03:13:44Z)
• LLDif: Diffusion Models for Low-light Emotion Recognition [15.095166627983566]
  This paper introduces LLDif, a novel diffusion-based facial expression recognition (FER) framework tailored for extremely low-light (LL) environments. Images captured under such conditions often suffer from low brightness and significantly reduced contrast, presenting challenges to conventional methods. LLDif addresses these issues with a novel two-stage training process that combines a Label-aware CLIP (LA-CLIP), an embedding prior network (PNET) and a transformer-based network adept at handling the noise of low-light images.
  arXiv  Detail & Related papers  (2024-08-08T05:41:09Z)
• AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancement [37.274077278901494]
  We propose the Attribute Guidance Diffusion framework (AGLLDiff) for effective real-world LIE. AGLLDiff shifts the paradigm and models the desired attributes, such as image exposure, structure and color of normal-light images. Our approach outperforms the current leading unsupervised LIE methods across benchmarks in terms of distortion-based and perceptual-based metrics.
  arXiv  Detail & Related papers  (2024-07-20T15:17:48Z)
• 2D Feature Distillation for Weakly- and Semi-Supervised 3D Semantic Segmentation [92.17700318483745]
  We propose an image-guidance network (IGNet) which builds upon the idea of distilling high level feature information from a domain adapted synthetically trained 2D semantic segmentation network. IGNet achieves state-of-the-art results for weakly-supervised LiDAR semantic segmentation on ScribbleKITTI, boasting up to 98% relative performance to fully supervised training with only 8% labeled points.
  arXiv  Detail & Related papers  (2023-11-27T07:57:29Z)
• Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Model [59.08821399652483]
  Illumination degradation image restoration (IDIR) techniques aim to improve the visibility of degraded images and mitigate the adverse effects of deteriorated illumination. Among these algorithms, diffusion model (DM)-based methods have shown promising performance but are often burdened by heavy computational demands and pixel misalignment issues when predicting the image-level distribution. We propose to leverage DM within a compact latent space to generate concise guidance priors and introduce a novel solution called Reti-Diff for the IDIR task. Reti-Diff comprises two key components: the Retinex-based latent DM (RLDM) and the Retinex-guided transformer (RG
  arXiv  Detail & Related papers  (2023-11-20T09:55:06Z)
• LLDiffusion: Learning Degradation Representations in Diffusion Models for Low-Light Image Enhancement [118.83316133601319]
  Current deep learning methods for low-light image enhancement (LLIE) typically rely on pixel-wise mapping learned from paired data. We propose a degradation-aware learning scheme for LLIE using diffusion models, which effectively integrates degradation and image priors into the diffusion process.
  arXiv  Detail & Related papers  (2023-07-27T07:22:51Z)

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.