Fool Me Once: Robust Selective Segmentation via Out-of-Distribution Detection with Contrastive Learning
This addresses the issue of high confidence in inaccurate segmentations for unknown classes in operational environments like driving scenes, though it is incremental.
The paper tackles the problem of unreliable segmentation in unknown scenes by training a network to perform segmentation and pixel-wise out-of-distribution detection, increasing segmentation accuracy by an IoU of 0.2 compared to alternative techniques.
In this work, we train a network to simultaneously perform segmentation and pixel-wise Out-of-Distribution (OoD) detection, such that the segmentation of unknown regions of scenes can be rejected. This is made possible by leveraging an OoD dataset with a novel contrastive objective and data augmentation scheme. By combining data including unknown classes in the training data, a more robust feature representation can be learned with known classes represented distinctly from those unknown. When presented with unknown classes or conditions, many current approaches for segmentation frequently exhibit high confidence in their inaccurate segmentations and cannot be trusted in many operational environments. We validate our system on a real-world dataset of unusual driving scenes, and show that by selectively segmenting scenes based on what is predicted as OoD, we can increase the segmentation accuracy by an IoU of 0.2 with respect to alternative techniques.