DeepLab
Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
Datos Bibliográficos
| ID | 23314935 |
|---|---|
| Autores | Liang-Chieh Chen (0000-0001-8564-374X, Google (United States)), George Papandreou (0000-0002-1468-2976, Google (United States)), Iasonas Kokkinos (0000-0002-2606-6476, University College London), Kevin Murphy (0000-0001-8982-3641, Google (United States)), Alan Yuille (0000-0001-5207-9249, Johns Hopkins University), Alan L Yuille |
| Año | 2018 |
| Volumen | 40 |
| Número | 4 |
| Páginas | 834-848 |
| Fecha de publicación | 2018-04-01 |
| Peer Reviewed | Sí |
| Open Access | Sí |
| Tipo | ARTICLE |
| Revista | IEEE Transactions on Pattern Analysis and Machine Intelligence (JOURNAL) |
| Identificadores de la revista | ISSN: 0162-8828 • E-ISSN: 1939-3539 |
| Editorial | Institute of Electrical and Electronics Engineers (IEEE) (PUBLISHER) |
| DOI | 10.1109/tpami.2017.2699184 |
| PMID | 28463186 |
| OpenAlex | W2412782625 |
| Idioma | EN |
| Citas recibidas | 89 |
| Referencias citadas | 64 |
In this work we address the task of semantic image segmentation with Deep Learning and make three main contributions that are experimentally shown to have substantial practical merit. First, we highlight convolution with upsampled filters, or 'atrous convolution', as a powerful tool in dense prediction tasks. Atrous convolution allows us to explicitly control the resolution at which feature responses are computed within Deep Convolutional Neural Networks. It also allows us to effectively enlarge the field of view of filters to incorporate larger context without increasing the number of parameters or the amount of computation. Second, we propose atrous spatial pyramid pooling (ASPP) to robustly segment objects at multiple scales. ASPP probes an incoming convolutional feature layer with filters at multiple sampling rates and effective fields-of-views, thus capturing objects as well as image context at multiple scales. Third, we improve the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models. The commonly deployed combination of max-pooling and downsampling in DCNNs achieves invariance but has a toll on localization accuracy. We overcome this by combining the responses at the final DCNN layer with a fully connected Conditional Random Field (CRF), which is shown both qualitatively and quantitatively to improve localization performance. Our proposed "DeepLab" system sets the new state-of-art at the PASCAL VOC-2012 semantic image segmentation task, reaching 79.7 percent mIOU in the test set, and advances the results on three other datasets: PASCAL-Context, PASCAL-Person-Part, and Cityscapes. All of our code is made publicly available online.
Artificial neural network · Conditional random field · Contextual image classification · Convolution (computer science) · Convolutional neural network · CRFS · Feature (linguistics) · Feature extraction · Image (mathematics) · Image segmentation · Markov random field · Object detection · Pascal (unit) · Pattern recognition (psychology) · Pooling · Segmentation · Upsampling · Advanced Image and Video Retrieval Techniques · Advanced Neural Network Applications · Artificial Intelligence · Computer Science · Robotics and Sensor-Based Localization
Jieb-linknet
FuNet
Obfuscated image classification for secure image-centric friend recommendation
Monitoring of Immovable Cultural Heritage Implementing 3D and Artificial Intelligence Technologies
Understanding cities with machine eyes
Measuring visual quality of street space and its temporal variation
Semantic Understanding of Scenes Through the ADE20K Dataset
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
Urban form and composition of street canyons
Detecting global urban expansion over the last three decades using a fully convolutional network
RCB-UNet
A Spatial Analysis of Urban Space Albedo According to Urban Surfaces Color
Extracting building footprints from UAV imagery for urban digital transformation in Ho Chi Minh City
Learning Frequency-Aware Cross-Modal Interaction for Multimodal Fake News Detection
MRI brain tumor segmentation using residual Spatial Pyramid Pooling-powered 3D U-Net
Assessing the impact of waterfront trail aesthetics on psychological restoration in urban environments
Comparative studies of deep learning segmentation models for left ventricle segmentation
Classification of Airborne Laser Scanning Point Cloud Using Point-Based Convolutional Neural Network
Multi-Supervised Feature Fusion Attention Network for Clouds and Shadows Detection
Dual Path Attention Net for Remote Sensing Semantic Image Segmentation
Cloud and Snow Segmentation in Satellite Images Using an Encoder–Decoder Deep Convolutional Neural Networks
A Research on Landslides Automatic Extraction Model Based on the Improved Mask R-CNN
High-Resolution Remote Sensing Image Segmentation Framework Based on Attention Mechanism and Adaptive Weighting
MCCRNet
Exploration of Semantic Geo-Object Recognition Based on the Scale Parameter Optimization Method for Remote Sensing Images
Urban Green Plastic Cover Mapping Based on VHR Remote Sensing Images and a Deep Semi-Supervised Learning Framework
Integration of Spatial and Co-Existence Relationships to Improve Administrative Region Target Detection in Map Images
HA-MPPNet
Micro-Terrain Recognition Method of Transmission Lines Based on Improved UNet
Assessing Similarities and Differences between Males and Females in Visual Behaviors in Spatial Orientation Tasks
Vepl-Net
GeoAI for Large-Scale Image Analysis and Machine Vision
Cross-Attention-Guided Feature Alignment Network for Road Crack Detection
Semantic Relation Model and Dataset for Remote Sensing Scene Understanding
A Modeling Method for Automatic Extraction of Offshore Aquaculture Zones Based on Semantic Segmentation
Crowd Density Estimation and Mapping Method Based on Surveillance Video and GIS
Convolutional Neural Network-Based Deep Learning Approach for Automatic Flood Mapping Using NovaSAR-1 and Sentinel-1 Data
Automatic Detection of Podotactile Pavements in Urban Environments Through a Deep Learning-Based Approach on MLS/HMLS Point Clouds
A Multi-Scale Water Extraction Convolutional Neural Network (MWEN) Method for GaoFen-1 Remote Sensing Images
Cascaded Residual Attention Enhanced Road Extraction from Remote Sensing Images
An Automatic Extraction Method for Hatched Residential Areas in Raster Maps Based on Multi-Scale Feature Fusion
Machine Learning of Spatial Data
MSA-UNet
Automated Processing of Remote Sensing Imagery Using Deep Semantic Segmentation
Exploration of an Open Vocabulary Model on Semantic Segmentation for Street Scene Imagery
Semantic Segmentation of High-Resolution Airborne Images with Dual-Stream DeepLabV3
Large-Scale Automatic Identification of Industrial Vacant Land
Classification of ground information during wildfires
Building Models to Assess Streetscape Quality Preferences Using Deep Learning Techniques
Ridesharing accessibility from the human eye
Assessing surface drainage conditions at the street and neighborhood scale
Examining the relationship between active transport and exposure to streetscape diversity during travel
A self-supervised detection method for mixed urban functions based on trajectory temporal image
Large-scale automatic identification of urban vacant land using semantic segmentation of high-resolution remote sensing images
Environmental determinants of dynamic jogging patterns
Utilizing street view images to estimate solar energy potential for photovoltaic-powered buses
Factors affecting the choices of adoption/non-adoption of future technologies during coronavirus pandemic
Evaluating building color harmoniousness in a historic district intelligently
Sensing climate justice
Mapping seasonal changes of street greenery using multi-temporal street-view images
Deep learning for optimizing urban governance by "sensing-processing-responding" cycle
Visual-language reasoning segmentation (Larse) of function-level building footprint across Yangtze River Economic Belt of China
Assessing the macro-scale patterns of urban tree canopy cover in Brazil using high-resolution remote sensing images
Towards Human-centric Digital Twins
Mapping urban green equity and analysing its impacted mechanisms
Greener is healthier
AdvUNet3
Street vitality and built environment features
Risk decision-making behavior in heritage tourism destinations based on eye-tracking experiments in Gulangyu, China
Quantifying Urban Surroundings Using Deep Learning Techniques
Deriving Environmental Properties Related to Human Environmental Perception
Social media insights into spatio-temporal emotional responses to Covid-19 crisis
How built environments shape leisure jogging styles
Integrating Street Views, Satellite Imageries and Remote Sensing Data Into Economics and the Social Sciences
Applying a mask R-CNN machine learning algorithm for segmenting electron microscope images of ceramic bronze-casting moulds
An object detection approach for detecting damages in heritage sites using 3-D point clouds and 2-D visual data
Understanding multi-perspective urban green space patterns under urban expansion
Unravelling the street thermal features network and spatial visual thermal inertia
A dual-system framework for rural spatial gentrification assessment using street view imagery
Unraveling nonlinear and interaction effects of multilevel built environment features on outdoor jogging with explainable machine learning
Extracting physical urban areas of 81 major Chinese cities from high-resolution land uses
How do disasters disrupt the spatial growth of informal settlements? A multi-temporal remote sensing approach – The case study of Mocoa, Colombia
Uncovering the edge urban villages using social media big data
Life between buildings from a street view image
Associating the Visual Built Environment with Active Transport
Computational Photography
Measuring the value of window views using real estate big data and computer vision
Fine-scale assessment of diurnal heat health risk based on satellite and street view images
Popularity influence mechanism of coastal spaces in urban areas
| Obras citantes distintas | 89 |
|---|---|
| Citas por año | 11,13 |
| Intervalo de citas | 2018 - 2026 (9) |
| Velocidad de citación | current |
| Altamente citado | No |
| Tipos de cita | Neutras: 84 |