Scailable. Edge AI Manager

Registration Details

Vision (switch)

Status

Service Status
Scailable API Online status unknown..
Module Online status unknown..
Daemon Online status unknown..
Output distributor Online status unknown..
Selected model
Number of cameras
Output location

Model settings

This will take you to an overview page with models and compute functions available for the current device. Select a model by clicking it. On the overview page you also have the option to add a Scailable model or compute function by uploading and transpiling a preexisting ONNX model.

Input settings

Camera 1

URL of image input source (http, https, file). The preview shows the full camera image, the model might not use the full area for input.

You can try one of the sample images

Select to use static or streaming input for the input URL.
Optional mask area. A JSON structure describing polygons that will be the mask for pixels to include in the detection.
When a mask is set only the part of the image within the mask is used for detection, the part outside of the mask is ignored.
Optional camera identification. A camera name makes it easier to identify the camera later.
Optional camera authentication If your camera needs authentication you can fill in the username and password.
Add camera

Camera sources input URL

You can use any IP camera as long as it has a accessible JPEG image endpoint or a MJPEG or RTSP streaming output.

The iSpy Connect database has a tool where you can look up a camera and create the URL to fill in.

Output settings

General output

The AI Manager sends results of the inference to this endpoint.

You can use the output distributor, which will locally store the data and automatically forward to the Scailable cloud. Data can be viewed on the device, and in the online dataviewer.

You can directly forward to the which will not store the data locally first. Data can be viewed in the online dataviewer.

You can also use a custom service like webhook.site or send data to other services with a REST interface like Elasticsearch or Node-RED.


The protocol used to transmit the inference results is REST by default. Other options like modbus are also possible, contact us for more details.

The formatting of the results is JSON by default.

Output Frequency

Inference results can be sent one at a time or aggregated.

  • If you aggregate the results over one input loop, the results for all input sources will be concatenated into one request to the logging server.
  • When you aggregate the inferences over time all results for all inputs will be combined into one request to the logging server.

Select a maximum frequency limit for the inferences in milliseconds, to save bandwidth and/or power. The model may not be fast enough to reach this maximum, but the module will not attempt faster inferences.

Upload images with low certainty

If the threshold is 0 no images will be sent to the API.
If the threshold is higher than 0 the input image will be sent to the API when an inference result has a probability value below the threshold.
If there are more detected classes, any detection in the result, independent of the detected class that is below the threshold will trigger that the image is sent to the API.

The AI Manager sends images with results that have low probability to this endpoint.
If empty the general output location will be used.

Test configuration

This will start the Scailable module, take one image for each enabled camera, feed image(s) to your selected model, print the output as it will be sent to the selected output location, display the image(s) in the camera preview field(s), and stop the module again.

Run model

This will minimize the configuration form, start the Scailable module, retrieve the input from each enabled camera, feed image(s) to your selected model, and send the result of the inferences to the specified output location. You can stop the model again by clicking on the stop button.

Status

Online status unknown.. Online status unknown.. Online status unknown..

Password protection

The AI manager can be protected with HTTP Basic Authentication.
You should enable this if your device is publicly accessible, e.g. by the exposed port 8081.

Be careful with this option.
When you select the checkbox, you must log in with the chosen password for user "scailable" immediately.
If you forgot the password the only way to reset it is to delete and reinstall the module.