Learning Instance Representation Banks for Aerial Scene Classification

Abstract

Aerial scenes are more complicated in terms of object distribution and spatial arrangement than natural scenes due to the bird view, and thus remain challenging to learn discriminative scene representation. Recent solutions design local semantic descriptors so that region of interests (RoIs) can be properly highlighted. However, each local descriptor has limited description capability and the overall scene representation remains to be refined. In this paper, we solve this problem by designing a novel representation set named instance representation bank (IRB), which unifies multiple local descriptors under the multiple instance learning (MIL) formulation. This unified framework is not trivial as all the local semantic descriptors can be aligned to the same scene scheme, enhancing the scene representation capability. Specifically, our IRB learning framework consists of a backbone, an instance representation bank, a semantic fusion module and a scene scheme alignment loss function. All the components are organized in an end-to-end manner. Extensive experiments on three aerial scene benchmarks demonstrate that our proposed method outperforms the state-of-the-art approaches by a large margin.

Classification of Scenes in Aerial Images with Deep Learning Models

“…For this reason, researchers in the literature have initially created benchmark data sets for artificial intelligence-based vision systems to be developed in this field [1][2][3][4][5]. Deep learning-based methods were developed on these datasets [6][7][8][9][10][11][12]. Deep learning is a sub-branch of artificial intelligence and first attracted attention with the ImageNet competition in 2012.…”

“…The AGOS method has been tested on three public datasets and has been reported to perform well. In [12], it is stated that objects in birds eye view images are more complex than objects in natural view, and therefore the discriminative features of scenes are difficult. To overcome this problem, a solution was sought by designing a new representation set called instance representation bank (IRB).…”