Multilevel Knowledge Transfer for Cross-Domain Object Detection
This addresses the problem of costly and infeasible annotated data acquisition for domain adaptation in object detection, offering an incremental improvement over existing methods.
The paper tackles the domain shift problem in object detection by proposing a multilevel knowledge transfer method that gradually reduces domain differences through pixel-level mapping, adversarial feature alignment, and pseudo-labeling, achieving significantly large performance gains over state-of-the-art approaches in challenging scenarios.
Domain shift is a well known problem where a model trained on a particular domain (source) does not perform well when exposed to samples from a different domain (target). Unsupervised methods that can adapt to domain shift are highly desirable as they allow effective utilization of the source data without requiring additional annotated training data from the target. Practically, obtaining sufficient amount of annotated data from the target domain can be both infeasible and extremely expensive. In this work, we address the domain shift problem for the object detection task. Our approach relies on gradually removing the domain shift between the source and the target domains. The key ingredients to our approach are -- (a) mapping the source to the target domain on pixel-level; (b) training a teacher network on the mapped source and the unannotated target domain using adversarial feature alignment; and (c) finally training a student network using the pseudo-labels obtained from the teacher. Experimentally, when tested on challenging scenarios involving domain shift, we consistently obtain significantly large performance gains over various recent state of the art approaches.