Focus-and-Detect: A Small Object Detection Framework for Aerial Images

TL;DR

This paper introduces a two-stage framework called 'Focus-and-Detect' for small object detection in aerial images, improving accuracy by focusing on regions of interest and employing a novel suppression method.

Contribution

The paper presents a novel two-stage detection framework with Gaussian Mixture Model supervision and Incomplete Box Suppression for improved small object detection in aerial images.

Findings

Achieved AP score of 42.06 on VisDrone dataset.

Outperformed existing small object detection methods.

Demonstrated effectiveness of focused region detection.

Abstract

Despite recent advances, object detection in aerial images is still a challenging task. Specific problems in aerial images makes the detection problem harder, such as small objects, densely packed objects, objects in different sizes and with different orientations. To address small object detection problem, we propose a two-stage object detection framework called "Focus-and-Detect". The first stage which consists of an object detector network supervised by a Gaussian Mixture Model, generates clusters of objects constituting the focused regions. The second stage, which is also an object detector network, predicts objects within the focal regions. Incomplete Box Suppression (IBS) method is also proposed to overcome the truncation effect of region search approach. Results indicate that the proposed two-stage framework achieves an AP score of 42.06 on VisDrone validation dataset, surpassing all other state-of-the-art small object detection methods reported in the literature, to the best of authors' knowledge.