In this article, we present the results of using Convolutional Auto-Encoders for de-noising raw data for CLAS12 drift chambers. The de-noising neural network provides increased efficiency in track reconstruction, also improved performance for high luminosity experimental data collection. The de-noising neural network used in conjunction with the previously developed track classifier neural network [1] lead to a significant track reconstruction efficiency increase for current luminosity (0 . 6 × 10 35 cm − 2 sec − 1 ). The increase in experimentally measured quantities will allow running experiments at twice the luminosity with the same track reconstruction efficiency. This will lead to huge savings in accelerator operational costs, and large savings for Jefferson Lab and collaborating institutions. we present a Machine Learning approach to de-noising detector data, the CLAS12 drift chambers specifically, using Convolutional Auto-Encoders. The data processed with the neural network and further processed with conventional tracking resulted in a significant increase in the number of reconstructed tracks. The study performed on simulated data shows a significant improvement in track reconstruction efficiency as a function of experimental luminosity. Using de-noising in combination with AI-assisted tracking further improves the track reconstruction efficiency. The resulting increase of physics events in MC is estimated to be 26% for three-particle final Staten the reaction H ( e, e (cid:48) π + π − ) p for the nominal experimental luminosity of CLAS12 (45 nA electron beam and a 5 cm long liquid hydrogen target). The efficiency of track reconstruction from the 95 nA beam background