Introduction to the model

This is an open-source object detection model by TensorFlow in TensorFlow Lite format.

This model was trained using the Common Objects in Context (COCO) dataset. COCO is a large-scale object detection dataset that is available for use under the Creative Commons Attribution 4.0 License. The dataset contains 80 object categories and 1.5 million object instances that include people, animals, food items, vehicles, and household items.

The model can be used in ArcGIS Survey123 to detect common objects in captured photos.

Note:

While it is not recommended to use this model in production surveys, it can be useful for demonstration purposes and to get started with smart assistants in Survey123. You are responsible for the use of this model. When using Survey123, it is your responsibility to review and manually correct its outputs.

License requirements

To complete this workflow, a Survey123 license is required. This model works with Survey123 Connect and the field app.

Model details

This model has the following characteristics:

• Input—A camera feed (either low-resolution preview or high-resolution capture) is used as input.
• Output—An image with common object detections written to its EXIF metadata or an image with redacted objects is output.
• Compute—This workflow is computationally intensive and performance may be impacted on some mobile devices.
• Architecture—An open source object detection model by TensorFlow in TensorFlow Lite format with MobileNet architecture.

Classes

You can limit the classes of objects that can be detected by specifying them using the class parameter. The class names are listed in the following table.

person	elephant	wine_glass	dining_table
bicycle	bear	cup	toilet
car	zebra	fork	tv
motorcycle	giraffe	knife	laptop
airplane	backpack	spoon	mouse
bus	umbrella	bowl	remote
train	handbag	banana	keyboard
truck	tie	apple	cell_phone
boat	suitcase	sandwich	microwave
traffic_light	frisbee	orange	oven
fire_hydrant	skis	broccoli	toaster
stop_sign	snowboard	carrot	sink
parking_meter	sports_ball	hot_dog	refrigerator
bench	kite	pizza	book
bird	baseball_bat	donut	clock
cat	baseball_glove	cake	vase
dog	skateboard	chair	scissors
horse	surfboard	couch	teddy_bear
sheep	tennis_racket	potted_plant	hair_dryer
cow	bottle	bed	toothbrush

Access and download the model

Download the Common Object Detection pretrained model from ArcGIS Living Atlas of the World. Alternatively, link the model to a survey in Survey123 Connect.

1. Browse to ArcGIS Living Atlas of the World.
2. Sign in with your ArcGIS Online credentials.
3. Search for Common Object Detection and open the item page from the search results.
4. Click the Download button to download the model.

   You can use the downloaded .dlpk file directly in Survey123 Connect by copying the file to the media folder for a survey.

Release notes

The following are the release notes:

Date	Description
February 2023	First release of Common Object Detection