Benchmark and Evaluation of Low Resource Object Detection in Biomedical Images

Object detection has seen tremendous progress in recent times, moving significantly closer to human-level performance. However, their performance is limited to domains with large amounts of annotated training data. A carefully curated, balanced data set is necessary for the successful training and deployment of current models. Their ability to adapt to novel domains, particularly those with significant visual variations such as biomedical domains, remains limited. Existing work on biomedical image analysis has focused on recognition with large amounts of training data. To address these limitations, we focus on creating a benchmark and evaluation platform for low-resource object detection, primarily for biomedical images. We introduce a novel data set to evaluate object detection models under resource-constrained environments such as biomedical images. We show that current models do not generalize to such significant domain change and provide a thorough analysis of failure modes to offer a way forward in this challenging task.