Low-Light Salient Object Detection by Learning to Highlight the Foreground Objects

Previous methods in salient object detection (SOD) mainly focused on favorable illumination circumstances while neglecting the performance in low-light condition, which significantly impedes the development of related down-stream tasks. In this work, considering that it is impractical to annotate the large-scale labels for this task, we present a framework (HDNet) to detect the salient objects in low-light images with the synthetic images. Our HDNet consists of a foreground highlight sub-network (HNet) and an appearance-aware detection sub-network (DNet), both of which can be learned jointly in an end-to-end manner. Specifically, to highlight the foreground objects, we design the HNet to estimate the parameters to adjust the dynamic range for each pixel adaptively, which can be trained via the weak supervision signals of the salient object labels. In addition, we design a simple detection network (DNet) with a contextual feature fusion module and a multi-scale feature refine module for detailed feature fusion and refinement. Furthermore, we contribute the first annotated dataset for salient object detection in low-light images (SOD-LL), including 6,000 labeled synthetic images (SOD-LLS) and 2,000 labeled real images (SOD-LLR). Experimental results on SOD-LL and other low-light videos in the wild demonstrate the effectiveness and generalization ability of our method. Our dataset and code are available at https://github.com/Ylinyuan/HDNet.

• Publication year

  2024

• Venue

  IEEE transactions on circuits and systems for video technology (Print)

• Publication date

  2024-08-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/TCSVT.2024.3377108
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Implicit Neural Representation for Cooperative Low-light Image Enhancement

  2023cited by this paper
• A Weakly Supervised Learning Framework for Salient Object Detection via Hybrid Labels

  2022cited by this paper
• URetinex-Net: Retinex-based Deep Unfolding Network for Low-light Image Enhancement

  2022cited by this paper
• I Can Find You! Boundary-Guided Separated Attention Network for Camouflaged Object Detection

  2022cited by this paper
• Multiscale Low-Light Image Enhancement Network With Illumination Constraint

  2022cited by this paper
• DHNet: Salient Object Detection With Dynamic Scale-Aware Learning and Hard-Sample Refinement

  2022cited by this paper
• Lightweight Salient Object Detection in Optical Remote Sensing Images via Feature Correlation

  2022cited by this paper
• Weakly-Supervised Salient Object Detection Using Point Supervison

  2022cited by this paper
• Low-Light Image Enhancement via Progressive-Recursive Network

  2021cited by this paper
• Dynamic Context-Sensitive Filtering Network for Video Salient Object Detection

  2021cited by this paper
• Dense Attentive Feature Enhancement for Salient Object Detection

  2021cited by this paper
• Complementary Trilateral Decoder for Fast and Accurate Salient Object Detection

  2021cited by this paper
• Pyramidal Feature Shrinking for Salient Object Detection

  2021cited by this paper
• Exploring Dense Context for Salient Object Detection

  2021cited by this paper
• Progressive Self-Guided Loss for Salient Object Detection

  2021influential reference
• Global Context-Aware Progressive Aggregation Network for Salient Object Detection

  2020cited by this paper
• Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancement

  2020cited by this paper
• Contour-Aware Loss: Boundary-Aware Learning for Salient Object Segmentation

  2020cited by this paper
• Exploring Rich and Efficient Spatial Temporal Interactions for Real-Time Video Salient Object Detection

  2020cited by this paper
• Exploring Image Enhancement for Salient Object Detection in Low Light Images

  2020cited by this paper
• Suppress and Balance: A Simple Gated Network for Salient Object Detection

  2020cited by this paper
• Label Decoupling Framework for Salient Object Detection

  2020cited by this paper
• Weakly-Supervised Salient Object Detection via Scribble Annotations

  2020cited by this paper
• Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement

  2020influential reference
• A Simple Pooling-Based Design for Real-Time Salient Object Detection

  2019cited by this paper
• LECARM: Low-Light Image Enhancement Using the Camera Response Model

  2019cited by this paper
• SAC-Net: Spatial Attenuation Context for Salient Object Detection

  2019cited by this paper
• Cascaded Partial Decoder for Fast and Accurate Salient Object Detection

  2019cited by this paper
• Salient Object Detection in the Deep Learning Era: An In-Depth Survey

  2019cited by this paper
• EnlightenGAN: Deep Light Enhancement Without Paired Supervision

  2019cited by this paper
• Underexposed Photo Enhancement Using Deep Illumination Estimation

  2019cited by this paper
• BASNet: Boundary-Aware Salient Object Detection

  2019cited by this paper
• Semi-Supervised Video Salient Object Detection Using Pseudo-Labels

  2019cited by this paper
• Stacked Cross Refinement Network for Edge-Aware Salient Object Detection

  2019cited by this paper
• Contour Knowledge Transfer for Salient Object Detection

  2018cited by this paper
• Enhanced-alignment Measure for Binary Foreground Map Evaluation

  2018cited by this paper
• Weakly Supervised Salient Object Detection Using Image Labels

  2018cited by this paper
• Learning to See in the Dark

  2018cited by this paper
• Getting to Know Low-light Images with The Exclusively Dark Dataset

  2018cited by this paper
• Non-local Deep Features for Salient Object Detection

  2017cited by this paper
• Structure-Measure: A New Way to Evaluate Foreground Maps

  2017cited by this paper
• Learning to Detect Salient Objects with Image-Level Supervision

  2017influential reference
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization

  2016cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

  2015cited by this paper
• Visual saliency based on multiscale deep features

  2015cited by this paper
• U-Net: Convolutional Networks for Biomedical Image Segmentation

  2015cited by this paper
• Saliency-aware geodesic video object segmentation

  2015cited by this paper
• LLNet: A deep autoencoder approach to natural low-light image enhancement

  2015cited by this paper
• STC: A Simple to Complex Framework for Weakly-Supervised Semantic Segmentation

  2015cited by this paper
• Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network

  2015cited by this paper
• Fully convolutional networks for semantic segmentation

  2014cited by this paper
• A Video Saliency Detection Model in Compressed Domain

  2014cited by this paper
• The Secrets of Salient Object Segmentation

  2014cited by this paper
• How to Evaluate Foreground Maps

  2014cited by this paper
• Saliency Detection via Graph-Based Manifold Ranking

  2013cited by this paper
• Hierarchical Saliency Detection

  2013cited by this paper
• Design and perceptual validation of performance measures for salient object segmentation

  2010cited by this paper
• Saliency detection using maximum symmetric surround

  2010cited by this paper
• Frequency-tuned salient region detection

  2009cited by this paper
• Learning to Detect A Salient Object

  2007cited by this paper
• Saliency Detection: A Spectral Residual Approach

  2007cited by this paper

• A Study on Real-time Object Detection using Deep Learning

  2026cites this paper
• TSNUNet: Two-Stage Nested U-Network for salient object detection

  2026cites this paper
• A comprehensive survey of visible and infrared imaging in complex environments: Principle, degradation and enhancement

  2025cites this paper
• EATNet: edge-aware and transformer-based network for RGB-D salient object detection

  2025cites this paper
• The Low-Illumination Catenary Component Detection Model Based on Semi-Supervised Learning and Adversarial Domain Adaptation

  2025cites this paper
• Semantic Prompt Enhancement for Semi-Supervised Low-Light Salient Object Detection

  2025cites this paper
• Low-light object detection via adaptive enhancement and dynamic feature fusion

  2025cites this paper
• A multi-scale vision mixture-of-experts for salient object detection with Kolmogorov-Arnold adapter

  2025cites this paper
• A low light object detection method based on a modified YOLO11 model

  2025cites this paper
• Dual-domain aware network for salient object detection in low-light images

  2025influential citation
• A Multi‐Scale Adaptive Network for Low‐Light Object Detection

  2025cites this paper
• A robust foreground detection method for local lighting changes via adaptive LBP and Gaussian mixture fusion

  2025cites this paper
• RDNet: Region Proportion-Aware Dynamic Adaptive Salient Object Detection Network in Optical Remote Sensing Images

  2025cites this paper
• Improved YOLOv11 for Low-illumination Object Detection in Autonomous Driving Scenarios

  2025cites this paper
• Salient Object Detection in Complex Weather Conditions via Noise Indicators

  2025cites this paper
• Breaking Barriers, Localizing Saliency: A Large-Scale Benchmark and Baseline for Condition-Constrained Salient Object Detection

  2025cites this paper
• Enhanced visual prompt meets low-light saliency detection

  2025cites this paper
• WXSOD: A Benchmark for Robust Salient Object Detection in Adverse Weather Conditions

  2025cites this paper
• Towards salient object detection via parallel dual-decoder network

  2025cites this paper
• An unsupervised low-light video enhancement network based on inter-frame consistency

  2024cites this paper
• Research on Low-Light Environment Object Detection Algorithm Based on YOLO_GD

  2024cites this paper