Contextual Object Detection Via Image Inpainting
Sinan Çavdar
Master Student
(Supervisor: Prof.Dr.Selim Aksoy) Computer Engineering Department
Bilkent University
Abstract: Object detection in aerial imagery is a critical task in computer vision with applications in urban monitoring, disaster management, and military surveillance. Current approaches often encounter challenges, such as a lack of representative features of objects with cluttered backgrounds. To address these challenges, we introduce CODI (Contextual Object Detection via Image Inpainting), a novel approach that extract the representative features of the image’s context derived from the inpainting model, and proposed fusion module inject the context into object detection model’s first stage for better localization and labeling accuracy over the objects. Our experiments were conducted on DOTA and HRSC2016 datasets. Our model achieved promising results on mAP by prevailing over Oriented RCNN, which is one of the best-performing models on the DOTA dataset. Specifically, by applying the same pre-processing and post-processing, CODI achieved an improvement of 0.67% in mAP (mean average precision) at single-scale and 0.6% at multi-scale over the Oriented RCNN on the test set. On the HRSC2016 dataset, CODI achieved an impressive mAP of 90.57%, outperforming Oriented RCNN by 0.27 percentage points. This result places CODI among the top-performing models using a ResNet-50 backbone. In that sense, our approach provides a foundation for future applications, paving the way for more precise object localization and identification in remote sensing.
DATE: July 08, Tuesday @ 13:00
PLACE: Zoom
This is an online seminar. To request event details please send a message to department.