Analyzing and Mitigating Bias for Vulnerable Classes: Towards Balanced Representation in Dataset
This work addresses fairness and reliability issues for vulnerable classes in autonomous driving, though it is incremental as it applies existing bias mitigation methods to a specific domain.
The paper tackled bias and class imbalances for vulnerable road users like cyclists and pedestrians in autonomous driving datasets, using CNN and Vision Transformer models on nuScenes data, and improved IoU and NDS metrics by up to 4.3 and 3.3 percentage points respectively with mitigation techniques like data augmentation and resampling.
The accuracy and fairness of perception systems in autonomous driving are essential, especially for vulnerable road users such as cyclists, pedestrians, and motorcyclists who face significant risks in urban driving environments. While mainstream research primarily enhances class performance metrics, the hidden traits of bias inheritance in the AI models, class imbalances and disparities within the datasets are often overlooked. Our research addresses these issues by investigating class imbalances among vulnerable road users, with a focus on analyzing class distribution, evaluating performance, and assessing bias impact. Utilizing popular CNN models and Vision Transformers (ViTs) with the nuScenes dataset, our performance evaluation indicates detection disparities for underrepresented classes. Compared to related work, we focus on metric-specific and Cost-Sensitive learning for model optimization and bias mitigation, which includes data augmentation and resampling. Using the proposed mitigation approaches, we see improvement in IoU(\%) and NDS(\%) metrics from 71.3 to 75.6 and 80.6 to 83.7 for the CNN model. Similarly, for ViT, we observe improvement in IoU and NDS metrics from 74.9 to 79.2 and 83.8 to 87.1. This research contributes to developing reliable models while enhancing inclusiveness for minority classes in datasets.