Efficient Multi-task Uncertainties for Joint Semantic Segmentation and Monocular Depth Estimation
This work addresses the need for efficient uncertainty quantification in autonomous driving applications, offering an incremental improvement over existing methods.
The paper tackled the problem of quantifying predictive uncertainty in multi-task learning for joint semantic segmentation and monocular depth estimation, introducing EMUFormer, which achieved state-of-the-art results on Cityscapes and NYUv2 datasets and produced high-quality uncertainties comparable to a Deep Ensemble while being significantly more efficient.
Quantifying the predictive uncertainty emerged as a possible solution to common challenges like overconfidence or lack of explainability and robustness of deep neural networks, albeit one that is often computationally expensive. Many real-world applications are multi-modal in nature and hence benefit from multi-task learning. In autonomous driving, for example, the joint solution of semantic segmentation and monocular depth estimation has proven to be valuable. In this work, we first combine different uncertainty quantification methods with joint semantic segmentation and monocular depth estimation and evaluate how they perform in comparison to each other. Additionally, we reveal the benefits of multi-task learning with regard to the uncertainty quality compared to solving both tasks separately. Based on these insights, we introduce EMUFormer, a novel student-teacher distillation approach for joint semantic segmentation and monocular depth estimation as well as efficient multi-task uncertainty quantification. By implicitly leveraging the predictive uncertainties of the teacher, EMUFormer achieves new state-of-the-art results on Cityscapes and NYUv2 and additionally estimates high-quality predictive uncertainties for both tasks that are comparable or superior to a Deep Ensemble despite being an order of magnitude more efficient.