UniT: Unified Knowledge Transfer for Any-shot Object Detection and Segmentation
[article]
Siddhesh Khandelwal, Raghav Goyal, Leonid Sigal
2021
arXiv
pre-print
Weakly-supervised approaches draw on image-level labels to build detectors/segmentors, while zero/few-shot methods assume abundant instance-level data for a set of base classes, and none to a few examples ...
Methods for object detection and segmentation rely on large scale instance-level annotations for training, which are difficult and time-consuming to collect. ...
annotations, ranging from no data (zero-shot) to a few (few-shot); (2) We propose a general, unified, interpretable and flexible end-to-end framework that can adopt classifiers/detectors/segmentors for ...
arXiv:2006.07502v3
fatcat:5wtstqlbcfbdplg2qzbpmmyq4a