Statements (148)
Predicate | Object |
---|---|
gptkbp:instance_of |
gptkb:Model
gptkb:television_channel |
gptkbp:applies_to |
Image Processing
semantic segmentation computer vision tasks |
gptkbp:architectural_style |
gptkb:Atrous_Convolution
|
gptkbp:based_on |
gptkb:television_channel
gptkb:Res_Net_architecture gptkb:Xception_architecture Fully Convolutional Networks (FCN) |
gptkbp:developed_by |
gptkb:Job_Search_Engine
gptkb:Google_Research |
gptkbp:enhances |
Feature Extraction
|
gptkbp:first_introduced |
gptkb:2016
|
gptkbp:first_released |
gptkb:2016
|
gptkbp:has_achievements |
state-of-the-art results
state-of-the-art performance State-of-the-art results Semantic Segmentation Performance |
gptkbp:has_variants |
gptkb:Deep_Labv1
gptkb:Deep_Labv2 gptkb:Deep_Labv3 gptkb:Deep_Labv3+ |
gptkbp:has_version |
gptkb:Deep_Lab_v1
gptkb:Deep_Lab_v2 gptkb:Deep_Lab_v3 gptkb:Deep_Lab_v3+ |
https://www.w3.org/2000/01/rdf-schema#label |
Deep Lab
|
gptkbp:improves |
Feature Extraction
Segmentation Accuracy image understanding image segmentation accuracy |
gptkbp:is_adopted_by |
Startups
Academic Institutions Industry Applications Tech Companies |
gptkbp:is_analyzed_in |
Visual Recognition
Image Segmentation Techniques Scene Parsing Urban Scene Understanding |
gptkbp:is_available_on |
gptkb:zoo
gptkb:Py_Torch_Hub gptkb:Tensor_Flow_Hub |
gptkbp:is_compatible_with |
gptkb:Graphics_Processing_Unit
gptkb:Py_Torch GPU acceleration multi-GPU training |
gptkbp:is_documented_in |
gptkb:archive
research papers academic papers technical reports Git Hub repositories |
gptkbp:is_evaluated_by |
gptkb:COCO_dataset
Cityscapes PASCALVOC Object Detection Tasks AD E20 K LIP dataset SBD dataset Instance Segmentation Tasks Segmentation Tasks |
gptkbp:is_implemented_in |
gptkb:Graphics_Processing_Unit
gptkb:Py_Torch |
gptkbp:is_incorporated_in |
Conditional Random Fields
Fully Connected Conditional Random Fields |
gptkbp:is_influenced_by |
gptkb:Alumni_Association
gptkb:Mask_R-CNN gptkb:Seg_Net Fully Convolutional Networks |
gptkbp:is_known_for |
high accuracy
real-time performance semantic segmentation pixel-wise classification adaptability to different domains contextual information utilization high-resolution segmentation robustness to occlusions |
gptkbp:is_open_source |
gptkb:theorem
|
gptkbp:is_optimized_for |
gptkb:smartphone
large-scale datasets GPU Processing |
gptkbp:is_part_of |
gptkb:viewpoint
gptkb:Artificial_Intelligence gptkb:Deep_Learning computer vision deep learning frameworks Deep Learning frameworks |
gptkbp:is_popular_in |
Computer Vision Community
|
gptkbp:is_recognized_for |
High Performance
Innovative Architecture Flexibility in Applications |
gptkbp:is_related_to |
image classification
object detection |
gptkbp:is_supported_by |
gptkb:software_framework
gptkb:document gptkb:Research_Institute Research Publications community contributions |
gptkbp:is_used_for |
Semantic Segmentation
|
gptkbp:is_used_in |
gptkb:Autonomous_Vehicles
gptkb:sports_team gptkb:Photographer gptkb:software gptkb:robot gptkb:metropolitan_area gptkb:brand Autonomous Driving agriculture Medical Imaging augmented reality autonomous driving e-commerce environmental monitoring gaming industry image editing artificial intelligence research video analysis facial recognition scene understanding image restoration social media applications agriculture technology object tracking medical image analysis security surveillance |
gptkbp:performance |
gptkb:Cityscapes_Dataset
gptkb:COCO_Dataset PASCALVOC AD E20 K Dataset |
gptkbp:provides |
pixel-level classification
Real-Time Segmentation |
gptkbp:published_by |
gptkb:ICCV_2019
gptkb:ECCV_2018 gptkb:CVPR_2017 |
gptkbp:requires |
GPU for training
|
gptkbp:successor |
gptkb:FCN_(Fully_Convolutional_Networks)
|
gptkbp:supports |
multi-scale prediction
multi-scale feature extraction Multi-Scale Contextual Information Multi-scale Contextual Information |
gptkbp:training |
large datasets
|
gptkbp:uses |
gptkb:Deep_Convolutional_Neural_Networks
convolutional neural networks |
gptkbp:utilizes |
gptkb:Atrous_Convolution
Conditional Random Fields atrous convolution |
gptkbp:bfsParent |
gptkb:television_channel
|
gptkbp:bfsLayer |
3
|