Grained Classification of Objects from Aerial Imagery | Robotics

Image credit: MAFAT

As the volume of imagery gathered by aerial sensors is rapidly growing, we understand that the exploitation of such data could not be achieved solely by a manual image analysis process. The competition’s objective is to explore automated solutions that will enable fine-grained classification of objects in high-resolution aerial imagery.

Participants goal is to detect and classify different objects found in high-resolution aerial imagery data. The classification includes fine-grained classification of sub-classes and unique features (Such as sunroof, color, air vents, etc.) Prizes:30,000 USD.

Source: CodaLab