Superpixel and brush tools.

[joaourbano]

Hi,

How can I access these tools?

Thanks

[ryouchinsa]

Thanks for using RectLabel.

You can find at the top bar Edit -> Label pixels menu to use brush and superpixel tools. https://rectlabel.com/help#label_pixels

[joaourbano]

Hi Ryo,

I understood from your your support page https://github.com/ryouchinsa/Rectlabel-support/blob/master/README.md that the automatic labeling of images is possible using Core ML from within rectLabel pro.

Cold you please provide me some guidance on this? It could a time saver for me, I am currently labeling around 10k pictures, where the automatic labeling of bottles, boxes and cans could be extremely useful.

Regards,

João

Sent from my iPad

On 22/12/2021, at 16:02, Ryo Kawamura @.***> wrote:

 Thanks for using RectLabel.

You can find at the top bar Edit -> Label pixels menu to use brush and superpixel tools. https://rectlabel.com/help#label_pixels

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you authored the thread.

[ryouchinsa]

Thanks for writing the issue.

If you are training your images/annotations on Create ML, the trained model is in Core ML format. https://rectlabel.com/help#xml_to_createml

Otherwise you have to convert your trained model to Core ML format. https://coremltools.readme.io/docs

RectLabel can load the Core ML model and automatically label images.

[joaourbano]

Hi Ryo,

I must have misunderstood something…

Indeed, my aim is to train a model in Create ML, and I am annotating images using RectLabel pro for that.

From your site https://github.com/ryouchinsa/Rectlabel-support/blob/master/README.md I got the following picture.

And my understanding was that rectLabel provided this functionality. Is this correct? And if yes, where can I download the model from?

Further to that on your site it says that “This blog provides MobileNetV2 + SSDLite with Core ML.” Where can download these pre-trained models?

Regards,

João

Sent from my iPad

On 07/01/2022, at 17:17, Ryo Kawamura @.***> wrote:

 Thanks for writing the issue.

If you are training your images/annotations on Create ML, the trained model is in Core ML format. https://rectlabel.com/help#xml_to_createml

Otherwise you have to convert your trained model to Core ML format. https://coremltools.readme.io/docs

RectLabel can load the Core ML model and automatically label images.

— Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android. You are receiving this because you authored the thread.

[ryouchinsa]

Hi,

You can use the Developer Tools menu in Xcode to open Create ML. https://developer.apple.com/machine-learning/create-ml/

Our document for loading core ml model is here. https://rectlabel.com/help#load_coreml

Apple provides Core ML Models for YOLOv3. https://developer.apple.com/machine-learning/models/

This blog provides MobileNetV2 + SSDLite with Core ML. https://machinethink.net/blog/mobilenet-ssdlite-coreml/

You can download the MobileNetV2 + SSDLite from here. https://github.com/hollance/coreml-survival-guide

[ryouchinsa]

Currently If you need our support to fix this problem, please let us know.