A deep neural network method for analyzing the CMS High Granularity Calorimeter (HGCAL) events

For the High Luminosity LHC, the CMS collaboration made the ambitious choice of a high granularity design to replace the existing endcap calorimeters. Thousands of particles coming from the multiple interactions create showers in the calorimeters, depositing energy simultaneously in adjacent cells. The data are similar to 3D gray-scale image that should be properly reconstructed. In this paper, we investigate how to localize and identify the thousands of showers in such events with a Deep Neural Network model. This problem is well-known in the “Vision” domain, it belongs to the challenging class: “Object Detection”. Our project shares a lot of similarities with the ones treated in Industry but faces several technological challenges like the 3D treatment. We present the Mask R-CNN model which has already proven its efficiency in Industry (for 2D images). We also present the first results and our plans to extend it to tackle 3D HGCAL data.