BugNIST - a Large Volumetric Dataset for Object Detection under Domain Shift

arXiv:2304.01838v2 Announce Type: replace
Abstract: Domain shift significantly influences the performance of deep learning algorithms, particularly for object detection within volumetric 3D images. Annotated training data is essential for deep learning-based object detection. However, annotating densely packed objects is time-consuming and costly. Instead, we suggest training models on individually scanned objects, causing a domain shift between training and detection data. To address this challenge, we introduce the BugNIST dataset, comprising 9154 micro-CT volumes of 12 bug types and 388 volumes …

abstract algorithms arxiv cs.ai cs.cv data dataset deep learning deep learning algorithms detection domain however images object objects performance shift training training data training models type