Detecting Objects in Aerial Images

Description

Object detection in Earth Vision, also known as Earth Observation and Remote Sensing, refers to the problem of localizing objects of interest (e.g., vehicles, airplanes and buildings) on the earth’s surface and predicting their corresponding categories. Observing plenty of instances from the overhead view provide a new way to understand the world. This is a relatively new field, with many new applications waiting to be developed. For movable categories, such as vehicles, ships, and planes, the orientation estimation is important for tracking. The majority of computer vision research focuses mostly on images from everyday life. However, the aerial imagery is a rich and structured source of information, yet, it is less investigated than it should be deserved. The task of object detection in aerial images is distinguished from the conventional object detection task in the following respects:

• The scale variations of object instances in aerial images are considerably huge.
• Many small object instances are densely distributed in aerial images, for example, the ships in a harbor and the vehicles in a parking lot.
• Objects in aerial images often appear in arbitrary orientations.

This workshop organizing on CVPR'2019, aims to draw attention from a wide range of communities and calls for more future research and efforts on the problems of object detection in aerial images. The workshop also contains a challenging on object detection in aerial images that features a new large-scale annotated image database of objects in aerial images, updated from DOTA-v1.0.

Topics

Topics of interests include, but are not limited to, following fields

• Object detection algorithms for optical (including multispectral and hyperspectral) remote sensing images
• Object detection algorithms for synthetic aperture radar (SAR) images
• Object detection algorithms and implementations for UAV platforms
• Deep learning models for object detection in aerial images
• Feature extraction for object detection in remote sensing images
• Benchmarks of object detection in remote sensing images
• Reviews and perspectives of object detection in remote sensing images
• Applications and systems of object detection in remote sensing images
• Object detection in Lidar point clouds

Submissions

Papers will be limited up to 8 pages, including figures and tables, according to the CVPR format (main conference authors’ guidelines). One can download the templates at LaTex/Word Templates(tar) or LaTex/Word Templates(zip). Papers will be reviewed by at least two reviewers with double blind policy. Papers will be selected based on their significance and novelty of results, technical merit, and clarity of presentation. Accepted papers will be published in CVPR 2019 proceedings and presented as posters in the workshop. Several papers will be selected as oral representation on the workshop. All the papers should be submitted through CMT website.