gptkbp:instance_of
|
gptkb:Model
|
gptkbp:applies_to
|
semantic segmentation
|
gptkbp:based_on
|
gptkb:television_channel
|
gptkbp:developed_by
|
gptkb:Google_Research
|
gptkbp:first_introduced
|
gptkb:2016
|
gptkbp:has_achievements
|
state-of-the-art performance
|
gptkbp:has_programs
|
augmented reality
satellite imagery analysis
urban scene understanding
|
https://www.w3.org/2000/01/rdf-schema#label
|
Deep Lab family of models
|
gptkbp:includes
|
gptkb:Deep_Lab_v1
gptkb:Deep_Lab_v2
gptkb:Deep_Lab_v3
gptkb:Deep_Lab_v3+
|
gptkbp:is_compatible_with
|
gptkb:Graphics_Processing_Unit
gptkb:Py_Torch
|
gptkbp:is_documented_in
|
gptkb:archive
tutorials
online courses
research articles
technical reports
|
gptkbp:is_evaluated_by
|
gptkb:COCO_dataset
gptkb:Cityscapes_dataset
AD E20 K dataset
PASCALVOC dataset
|
gptkbp:is_influenced_by
|
gptkb:Alumni_Association
gptkb:Seg_Net
Fully Convolutional Networks
|
gptkbp:is_open_source
|
gptkb:theorem
|
gptkbp:is_optimized_for
|
real-time inference
|
gptkbp:is_part_of
|
deep learning community
computer vision field
|
gptkbp:is_scalable
|
high-resolution images
|
gptkbp:is_supported_by
|
community contributions
|
gptkbp:is_used_by
|
gptkb:physicist
gptkb:software
data scientists
|
gptkbp:is_used_in
|
gptkb:robot
autonomous driving
medical image analysis
|
gptkbp:performance
|
pixel accuracy
mean Intersection over Union (m Io U)
|
gptkbp:publishes
|
gptkb:Encoder-Decoder_with_Atrous_Separable_Convolution_for_Semantic_Image_Segmentation
Deep Lab v3: Rethinking Atrous Convolution for Semantic Image Segmentation
Deep Lab v3+: Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
Deep Lab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Conditional Random Fields
|
gptkbp:requires
|
large memory capacity
GPU for training
|
gptkbp:security_features
|
image noise
occlusions
lighting variations
|
gptkbp:supports
|
multi-scale feature extraction
|
gptkbp:training
|
large-scale datasets
|
gptkbp:uses
|
gptkb:Atrous_Convolution
|
gptkbp:utilizes
|
Conditional Random Fields
|
gptkbp:bfsParent
|
gptkb:Deep_Lab_v3+
|
gptkbp:bfsLayer
|
5
|