Statements (147)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:cosmic_ray_detector
gptkb:AI_technology |
| gptkbp:applies_to |
transfer learning
computer vision tasks multi-scale context |
| gptkbp:based_on |
gptkb:neural_networks
gptkb:Deep_Lab_v3 |
| gptkbp:can |
objects in images
|
| gptkbp:can_be_fine_tuned_for |
specific tasks
|
| gptkbp:can_be_fine-tuned_for |
specific tasks
|
| gptkbp:can_be_used_in |
image editing
object detection video analysis scene understanding |
| gptkbp:can_handle |
variable input sizes
|
| gptkbp:deployment |
gptkb:mobile_applications
|
| gptkbp:developed_by |
gptkb:Google
gptkb:Google_Research |
| gptkbp:enhances |
object boundary detection
|
| gptkbp:features |
atrous separable convolution
|
| gptkbp:has |
pre-trained models
open-source code |
| gptkbp:has_achieved |
state-of-the-art results
state-of-the-art performance |
| https://www.w3.org/2000/01/rdf-schema#label |
Deep Lab v3+
|
| gptkbp:improves |
gptkb:Deep_Lab_v3
semantic segmentation semantic segmentation accuracy |
| gptkbp:includes |
gptkb:ASPP_module
|
| gptkbp:input_output |
pixel-wise classification
pixel-wise predictions |
| gptkbp:introduced |
atrous convolution
|
| gptkbp:introduced_in |
gptkb:2018
|
| gptkbp:is_applicable_to |
autonomous driving
satellite imagery analysis medical image analysis |
| gptkbp:is_available_on |
gptkb:Git_Hub
|
| gptkbp:is_based_on |
Deep Lab framework
|
| gptkbp:is_compatible_with |
gptkb:NVIDIA_GPUs
gptkb:TPUs cloud platforms various hardware platforms |
| gptkbp:is_considered_as |
a leading model in segmentation tasks
|
| gptkbp:is_designed_for |
dense prediction tasks
|
| gptkbp:is_documented_in |
research papers
technical blogs Git Hub repositories |
| gptkbp:is_evaluated_by |
gptkb:Open_Images_dataset
gptkb:Image_Net gptkb:PASCAL_VOC_dataset gptkb:COCO_dataset gptkb:Cam_Vid_dataset gptkb:Cityscapes_dataset gptkb:KITTI_dataset gptkb:ADE20_K_dataset industry professionals academic researchers pixel accuracy SUN dataset LIP dataset mean Intersection over Union (m Io U) SBD dataset mean Intersection over Union (m Io U) metric |
| gptkbp:is_implemented_in |
gptkb:Tensor_Flow
gptkb:Py_Torch |
| gptkbp:is_influenced_by |
gptkb:Res_Net
gptkb:Xception gptkb:Res_Net_architecture gptkb:Xception_architecture |
| gptkbp:is_integrated_with |
other neural networks
other AI models |
| gptkbp:is_known_for |
high accuracy
user-friendly interfaces real-time performance efficient computation flexibility in applications high-resolution outputs flexibility in architecture high resolution outputs real-time inference capabilities robustness to occlusions ability to handle varying object scales |
| gptkbp:is_open_source |
gptkb:true
|
| gptkbp:is_optimized_for |
GPU acceleration
|
| gptkbp:is_part_of |
gptkb:Deep_Lab_family_of_models
deep learning frameworks AI research initiatives AI research community computer vision research AI toolkits computer vision frameworks |
| gptkbp:is_popular_in |
gptkb:scientific_community
computer vision community |
| gptkbp:is_recognized_for |
contribution to semantic segmentation.
|
| gptkbp:is_related_to |
gptkb:Artificial_Intelligence
gptkb:machine_learning image segmentation |
| gptkbp:is_scalable |
large datasets
large images |
| gptkbp:is_supported_by |
gptkb:Py_Torch_Hub
gptkb:Tensor_Flow_Hub academic institutions community contributions |
| gptkbp:is_tested_for |
real-world applications
benchmark datasets |
| gptkbp:is_trained_in |
gptkb:PASCAL_VOC_dataset
gptkb:COCO_dataset gptkb:Cityscapes_dataset gptkb:Pascal_VOC_dataset large datasets |
| gptkbp:is_used_for |
object detection
image analysis object detection tasks |
| gptkbp:is_used_in |
gptkb:robotics
augmented reality autonomous driving environmental monitoring autonomous driving systems virtual reality applications video analysis smart city projects robotics applications security applications augmented reality applications agriculture technology fashion analysis medical image analysis image segmentation tasks image segmentation competitions |
| gptkbp:performance |
image segmentation tasks
|
| gptkbp:released_in |
gptkb:2018
|
| gptkbp:requires |
GPU for training
large datasets for training |
| gptkbp:security |
image noise
|
| gptkbp:suitable_for |
gptkb:mobile_applications
|
| gptkbp:supports |
real-time processing
multi-scale feature extraction multi-scale context |
| gptkbp:training |
backpropagation
|
| gptkbp:used_for |
semantic segmentation
|
| gptkbp:uses |
gptkb:feature_pyramid_networks
encoder-decoder architecture atrous convolution |
| gptkbp:utilizes |
encoder-decoder architecture
atrous convolution |
| gptkbp:bfsParent |
gptkb:Deep_Lab
|
| gptkbp:bfsLayer |
5
|