The integrated circuit (IC) delayering workflow is heavily reliant on operator experience to determine the processing end point, which is the ideal point on an IC where processing should be terminated, to optimize region of interest imaging. The current method of end point detection during IC delayering utilizes qualitative correlation between dielectric film color and dielectric thickness observed via optical microscopy to guide decision making. The goal of this work is to quantify this relationship using computer vision. In the field of computer vision, convolutional neural networks (CNNs) have been successfully applied to capture spatial relationships within images. Given this success, a CNN was trained for thickness estimates of dielectric films using optical images captured during processing for eventual automated end point detection. The trained model explained 39% of the variance in dielectric film thickness with a mean absolute error of approximately 47 nm.