Deep learning models

Доступно с лицензией Image Analyst.

The following table provides an overview of the deep learning models available in ArcGIS Pro. Each row provides compatible metadata formats and the main use of the specific model. Where available, accompanying examples are included.

Deep learning model typeSupported metadataTaskExample

BDCN Edge Detector External link

Classified tiles

Pixel classification

Land parcel extraction External link

Change Dectector External link

Classified tiles

Change detection

Pixel classification (change detection)

Detect building changeExternal link

ConnectNet

Classified tiles

Pixel classification

CycleGAN External link

Export tiles

CycleGAN

Image translation (unpaired images)

SAR to RGB translationExternal link

DeepLab External link

Classified tiles

Pixel classification

Faster RCNN External link

PASCAL_VOC_rectangles

KITTI_rectangles

Object detection

Feature classifier External link

Labeled tiles

Imagenet

Multi-labeled tiles

Object detection

Feature categorization External link

HED Edge Detector External link

Classified tiles

Pixel classification

Land parcel extraction External link

Image CaptionerExternal link

Image captioning

Image translation

Mask RCNN External link

RCNN masks

Object detection (instance segmentation)

Caribou detection and classification External link

MMDetection

PASCAL_VOC_rectangles

KITTI_rectangles

Object detection

MMSegentation

Classified tiles

Pixel classification

Multi-Task Road Extractor External link

Classified tiles

Pixel classification

Automatic road extraction External link

Pix2Pix External link

Export tiles

Image translation (paired imges)

Colorizing historic imagery External link

PSPNet External link

Classified tiles

Pixel classification

RetinaNet External link

PASCAL_VOC_rectangles

KITTI_rectangles

Object detection

Utility and vegetation detection External link

Siam Mask External link

RCNN Masks

Object tracker

SSD External link

PASCAL_VOC_rectangles

KITTI_rectangles

Object detection

Detect palm tree health External link

Super-resolution External link

Super-resolution

Image translation

Increase image resolutionExternal link

U-Net External link

Classified tiles

Pixel classification

Extracting building footprintsExternal link

YOLOv3 External link

PASCAL_VOC_rectangles

KITTI_rectangles

Object detection

Примечание:

Some of the examples that use the Python Notebook for training can be performed using the Train Deep Learning Model tool.

Связанные разделы