U-Net: Going Deeper with Nested U-Structure for Salient Object Detection
arXiv:2005.09007 · doi:10.1016/j.patcog.2020.107404
Abstract
In this paper, we design a simple yet powerful deep network architecture, U-Net, for salient object detection (SOD). The architecture of our U-Net is a two-level nested U-structure. The design has the following advantages: (1) it is able to capture more contextual information from different scales thanks to the mixture of receptive fields of different sizes in our proposed ReSidual U-blocks (RSU), (2) it increases the depth of the whole architecture without significantly increasing the computational cost because of the pooling operations used in these RSU blocks. This architecture enables us to train a deep network from scratch without using backbones from image classification tasks. We instantiate two models of the proposed architecture, U-Net (176.3 MB, 30 FPS on GTX 1080Ti GPU) and U-Net (4.7 MB, 40 FPS), to facilitate the usage in different environments. Both models achieve competitive performance on six SOD datasets. The code is available: https://github.com/NathanUA/U-2-Net.
Accepted in Pattern Recognition 2020
References in corpus (2)
Cited by in corpus (98)
- UIU-Net: U-Net in U-Net for Infrared Small Object Detection
- Multi-Attention-Network for Semantic Segmentation of Fine Resolution Remote Sensing Images
- A2-FPN for Semantic Segmentation of Fine-Resolution Remotely Sensed Images
- Fast Camouflaged Object Detection via Edge-based Reversible Re-calibration Network
- Automatic Sleep Staging of EEG Signals: Recent Development, Challenges, and Future Directions
- MINet: Multi-scale Interactive Network for Real-time Salient Object Detection of Strip Steel Surface Defects
- Multi-Content Complementation Network for Salient Object Detection in Optical Remote Sensing Images
- TransAttUnet: Multi-level Attention-guided U-Net with Transformer for Medical Image Segmentation
- SAM2-UNet: Segment Anything 2 Makes Strong Encoder for Natural and Medical Image Segmentation
- MFRNet: A New CNN Architecture for Post-Processing and In-loop Filtering
- Bilateral Reference for High-Resolution Dichotomous Image Segmentation
- Boundary-Aware Segmentation Network for Mobile and Web Applications
- CarDD: A New Dataset for Vision-based Car Damage Detection
- Domain adaptation based self-correction model for COVID-19 infection segmentation in CT images
- PaddleSeg: A High-Efficient Development Toolkit for Image Segmentation
- Free-form tumor synthesis in computed tomography images via richer generative adversarial network
- Quality-aware Selective Fusion Network for V-D-T Salient Object Detection
- Salient Objects in Clutter
- Development of Skip Connection in Deep Neural Networks for Computer Vision and Medical Image Analysis: A Survey
- Inter- and intra-uncertainty based feature aggregation model for semi-supervised histopathology image segmentation
- Counterfactual Generative Networks
- Inkspire: Supporting Design Exploration with Generative AI through Analogical Sketching
- Robust Saliency-Aware Distillation for Few-shot Fine-grained Visual Recognition
- SFDFusion: An Efficient Spatial-Frequency Domain Fusion Network for Infrared and Visible Image Fusion
- CSC-Unet: A Novel Convolutional Sparse Coding Strategy Based Neural Network for Semantic Segmentation
- A modular U-Net for automated segmentation of X-ray tomography images in composite materials
- Spectrum-driven Mixed-frequency Network for Hyperspectral Salient Object Detection
- RNDiff: Rainfall nowcasting with Condition Diffusion Model
- Salient Mask-Guided Vision Transformer for Fine-Grained Classification
- SKA Science Data Challenge 2: analysis and results
- Saliency-Aware Spatio-Temporal Artifact Detection for Compressed Video Quality Assessment
- Hybrid Skip: A Biologically Inspired Skip Connection for the UNet Architecture
- Towards Interactive Image Inpainting via Sketch Refinement
- Trichomonas Vaginalis Segmentation in Microscope Images
- Res-U2Net: Untrained Deep Learning for Phase Retrieval and Image Reconstruction
- Self-Normalized Density Map (SNDM) for Counting Microbiological Objects
- A Two-Stage Imaging Framework Combining CNN and Physics-Informed Neural Networks for Full-Inverse Tomography: A Case Study in Electrical Impedance Tomography (EIT)
- DocScanner: Robust Document Image Rectification with Progressive Learning
- Glance and Gaze: A Collaborative Learning Framework for Single-channel Speech Enhancement
- SalientSleepNet: Multimodal Salient Wave Detection Network for Sleep Staging
- Domain Adaptive Detection of MAVs: A Benchmark and Noise Suppression Network
- Continuous Gaze Tracking With Implicit Saliency-Aware Calibration on Mobile Devices
- The Impact of Background Removal on Performance of Neural Networks for Fashion Image Classification and Segmentation
- Perceptual Conversational Head Generation with Regularized Driver and Enhanced Renderer
- Multitask Learning in Minimally Invasive Surgical Vision: A Review
- Relightable 3D Head Portraits from a Smartphone Video
- Ab Initio Particle-based Object Manipulation
- Weakly-supervised Instance Segmentation via Class-agnostic Learning with Salient Images
- Can You Spot the Chameleon? Adversarially Camouflaging Images from Co-Salient Object Detection
- Scrape, Cut, Paste and Learn: Automated Dataset Generation Applied to Parcel Logistics
- Retinal OCT Synthesis with Denoising Diffusion Probabilistic Models for Layer Segmentation
- DualStreamFoveaNet: A Dual Stream Fusion Architecture with Anatomical Awareness for Robust Fovea Localization
- A Physics-Embedded Dual-Learning Imaging Framework for Electrical Impedance Tomography
- PSGR: Pixel-wise Sparse Graph Reasoning for COVID-19 Pneumonia Segmentation in CT Images
- JOKR: Joint Keypoint Representation for Unsupervised Cross-Domain Motion Retargeting
- Comprehensive Saliency Fusion for Object Co-segmentation
- MixSA: Training-free Reference-based Sketch Extraction via Mixture-of-Self-Attention
- Rethinking the Nested U-Net Approach: Enhancing Biomarker Segmentation with Attention Mechanisms and Multiscale Feature Fusion
- BiconNet: An Edge-preserved Connectivity-based Approach for Salient Object Detection
- Grounding inductive biases in natural images:invariance stems from variations in data
- Guiding Attention in End-to-End Driving Models
- Object Pose Estimation by Camera Arm Control Based on the Next Viewpoint Estimation
- ICDAR 2021 Competition on Components Segmentation Task of Document Photos
- Salience-guided Ground Factor for Robust Localization of Delivery Robots in Complex Urban Environments
- SLIDE: Single Image 3D Photography with Soft Layering and Depth-aware Inpainting
- Glissando-Net: Deep sinGLe vIew category level poSe eStimation ANd 3D recOnstruction
- Looking for change? Roll the Dice and demand Attention
- Exploring Saliency Bias in Manipulation Detection
- U2-ONet: A Two-level Nested Octave U-structure with Multiscale Attention Mechanism for Moving Instances Segmentation
- ReTiDe: Real-Time Denoising for Energy-Efficient Motion Picture Processing with FPGAs
- UruDendro4: A Benchmark Dataset for Automatic Tree-Ring Detection in Cross-Section Images of Pinus taeda L
- ResNEsts and DenseNEsts: Block-based DNN Models with Improved Representation Guarantees
- PL-Net: Progressive Learning Network for Medical Image Segmentation
- DC-GNet: Deep Mesh Relation Capturing Graph Convolution Network for 3D Human Shape Reconstruction
- Video Frame Interpolation via Structure-Motion based Iterative Fusion
- Receptive Field Broadening and Boosting for Salient Object Detection
- Defocus Blur Detection via Salient Region Detection Prior
- Salient Image Matting
- Multi-scale Edge-based U-shape Network for Salient Object Detection
- Fairness is in the details: Face Dataset Auditing
- Fully Understanding Generic Objects: Modeling, Segmentation, and Reconstruction
- Learning to Segment Rigid Motions from Two Frames
- FLIM Networks with Bag of Feature Points
- Improving Building Segmentation for Off-Nadir Satellite Imagery
- Deep Learning Approach Protecting Privacy in Camera-Based Critical Applications
- Appearance Editing with Free-viewpoint Neural Rendering
- MNIST-Gen: A Modular MNIST-Style Dataset Generation Using Hierarchical Semantics, Reinforcement Learning, and Category Theory
- JIT-Masker: Efficient Online Distillation for Background Matting
- Cascade Image Matting with Deformable Graph Refinement
- D-Judge: How Far Are We? Assessing the Discrepancies Between AI-synthesized and Natural Images through Multimodal Guidance
- Characterizing and Improving the Robustness of Self-Supervised Learning through Background Augmentations
- MTFusion: Reconstructing Any 3D Object from Single Image Using Multi-word Textual Inversion
- Dual-Context Aggregation for Universal Image Matting
- Analyzing Adversarial Robustness of Deep Neural Networks in Pixel Space: a Semantic Perspective
- Face Sketch Synthesis via Semantic-Driven Generative Adversarial Network
- HistoSeg++: Delving deeper with attention and multiscale feature fusion for biomarker segmentation
- Deep Automatic Natural Image Matting
- A General Divergence Modeling Strategy for Salient Object Detection