You recently built the first version of an image segmentation model for a self-driving car, which is meant to identify and separate different objects in traffic scenes from video input. After deploying the model, you observe a decrease in the area under the curve (AUC) metric, indicating a drop in performance. When analyzing the video recordings, you discover that the model performs well in scenarios with less traffic but fails in highly congested traffic conditions. What is the most likely reason for this performance issue?