LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation

Abstract

Pixel-wise semantic segmentation for visual scene understanding not only needs to be accurate, but also efficient in order to find any use in real-time application. Existing algorithms even though are accurate but they do not focus on utilizing the parameters of neural network efficiently. As a result they are huge in terms of parameters and number of operations; hence slow too. In this paper, we propose a novel deep neural network architecture which allows it to learn without any significant increase in number of parameters. Our network uses only 11.5 million parameters and 21.2 GFLOPs for processing an image of resolution 3x640x360. It gives state-of-the-art performance on CamVid and comparable results on Cityscapes dataset. We also compare our networks processing time on NVIDIA GPU and embedded system device with existing state-of-the-art architectures for different image resolutions.

5 pages, 5 figures, GitHub: https://github.com/e-lab/LinkNet

References in corpus (1)

• Conditional Random Fields as Recurrent Neural Networks

Cited by in corpus (40)

• Landslide4Sense: Reference Benchmark Data and Deep Learning Models for Landslide Detection
• Using DUCK-Net for Polyp Image Segmentation
• Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical Segmentation
• DDU-Net: Dual-Decoder-U-Net for Road Extraction Using High-Resolution Remote Sensing Images
• R2U++: A Multiscale Recurrent Residual U-Net with Dense Skip Connections for Medical Image Segmentation
• Real-Time Instrument Segmentation in Robotic Surgery using Auxiliary Supervised Deep Adversarial Learning
• Universal Adversarial Examples in Remote Sensing: Methodology and Benchmark
• EndoMapper dataset of complete calibrated endoscopy procedures
• Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision
• MP-ResNet: Multi-path Residual Network for the Semantic segmentation of High-Resolution PolSAR Images
• VecQ: Minimal Loss DNN Model Compression With Vectorized Weight Quantization
• Global-Reasoned Multi-Task Learning Model for Surgical Scene Understanding
• Bounding Box-Free Instance Segmentation Using Semi-Supervised Learning for Generating a City-Scale Vehicle Dataset
• Road Extraction with Satellite Images and Partial Road Maps
• AP-MTL: Attention Pruned Multi-task Learning Model for Real-time Instrument Detection and Segmentation in Robot-assisted Surgery
• Panoptic Instance Segmentation on Pigs
• Multi-scale Cloud Detection in Remote Sensing Images using a Dual Convolutional Neural Network
• Task Decomposition and Synchronization for Semantic Biomedical Image Segmentation
• Relict landslide detection using Deep-Learning architectures for image segmentation in rainforest areas: A new framework
• PaveSAM Segment Anything for Pavement Distress
• Backdoor Attacks for Remote Sensing Data with Wavelet Transform
• Context-Integrated and Feature-Refined Network for Lightweight Object Parsing
• Deep Neural Networks for Automatic Grain-matrix Segmentation in Plane and Cross-polarized Sandstone Photomicrographs
• Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection
• Automatic Diagnosis of Pneumothorax from Chest Radiographs: A Systematic Literature Review
• Semantic Segmentation of Histopathological Slides for the Classification of Cutaneous Lymphoma and Eczema
• SFSegNet: Parse Freehand Sketches using Deep Fully Convolutional Networks
• A Comparative Analysis of CNN-based Deep Learning Models for Landslide Detection
• A Deep Learning Approach to Teeth Segmentation and Orientation from Panoramic X-rays
• CNN-Based Semantic Change Detection in Satellite Imagery
• FAS-UNet: A Novel FAS-driven Unet to Learn Variational Image Segmentation
• A Fair Evaluation of Various Deep Learning-Based Document Image Binarization Approaches
• Progressive Cross Attention Network for Flood Segmentation using Multispectral Satellite Imagery
• A study of CNN capacity applied to Left Venticle Segmentation in Cardiac MRI
• Learning to Segment Human Body Parts with Synthetically Trained Deep Convolutional Networks
• Synthetic Data-Driven Multi-Architecture Framework for Automated Polyp Segmentation Through Integrated Detection and Mask Generation
• Low to High Dimensional Modality Hallucination using Aggregated Fields of View
• TSLA: A Task-Specific Learning Adaptation for Semantic Segmentation on Autonomous Vehicles Platform
• Drawing the Line: Deep Segmentation for Extracting Art from Ancient Etruscan Mirrors
• Automated Charge Transition Detection in Quantum Dot Charge Stability Diagrams