Hi,

I am looking into the possibility of using RDA to detect concrete trucks on construction sites and also to detect PPE on site. (These would probably be two separate models and so Concrete trucks is the one I would like to focus on.) I have been going through Reality Data Analysis API: https://developer.bentley.com/apis/realitydataanalysis/tutorials/ and found the tutorial: https://developer.bentley.com/tutorials/realitydataanalysis-api-quick-start/. I realised I would need a specific dataset for this after looking at: https://communities.bentley.com/products/3d_imaging_and_point_cloud_software/w/wiki/54656/context-insights-detectors-download-page. I also noted that this feature is just a technological preview at the moment.

How is the best way I can prepare for this feature for the future? For example I could begin collecting the require images and labels. So how many images would be needed, types of images, how should the data be labelled, xml files? Thank you for any help you can provide with this.

Kind regards, Brinley