A Study on Quantifying Sim2Real Image Gap in Autonomous Driving Simulations Using Lane Segmentation Attention Map Similarity
This addresses the problem of quantifying image realism for researchers and developers in autonomous driving simulations, but it is incremental as it builds on existing ideas like FID.
The paper tackles the lack of a quantitative metric for evaluating the realism of simulation images in autonomous driving by proposing a new metric based on attention map similarity from ENet-SAD, and verifies it on CARLA simulator images.
Autonomous driving simulations require highly realistic images. Our preliminary study found that when the CARLA Simulator image was made more like reality by using DCLGAN, the performance of the lane recognition model improved to levels comparable to real-world driving. It was also confirmed that the vehicle's ability to return to the center of the lane after deviating from it improved significantly. However, there is currently no agreed-upon metric for quantitatively evaluating the realism of simulation images. To address this issue, based on the idea that FID (Fréchet Inception Distance) measures the feature vector distribution distance using a pre-trained model, this paper proposes a metric that measures the similarity of simulation road images using the attention map from the self-attention distillation process of ENet-SAD. Finally, this paper verified the suitability of the measurement method by applying it to the image of the CARLA map that implemented a realworld autonomous driving test road.