Probing Contextual Diversity for Dense Out-of-Distribution Detection
This work addresses the challenge of dense OoD detection for semantic segmentation tasks, which is incremental as it builds on existing OoD detection methods by focusing on context.
The paper tackles the problem of out-of-distribution (OoD) segmentation in image analysis, which is less studied than OoD classification, by introducing MOoSe, a method that aggregates multi-scale contextual representations to improve detection and uncertainty estimation, showing consistently positive effects.
Detection of out-of-distribution (OoD) samples in the context of image classification has recently become an area of interest and active study, along with the topic of uncertainty estimation, to which it is closely related. In this paper we explore the task of OoD segmentation, which has been studied less than its classification counterpart and presents additional challenges. Segmentation is a dense prediction task for which the model's outcome for each pixel depends on its surroundings. The receptive field and the reliance on context play a role for distinguishing different classes and, correspondingly, for spotting OoD entities. We introduce MOoSe, an efficient strategy to leverage the various levels of context represented within semantic segmentation models and show that even a simple aggregation of multi-scale representations has consistently positive effects on OoD detection and uncertainty estimation.