Sim2Air - Synthetic aerial dataset for UAV monitoring
This work addresses the challenge of improving object detection in UAV monitoring under conditions like poor illumination and small object sizes, representing an incremental advance in synthetic data generation for domain adaptation.
The paper tackled the problem of generating synthetic aerial datasets for UAV monitoring by applying texture randomization to accentuate shape-based object representation, resulting in increased mAP values by up to 20 percentage points for object detectors like YOLO and Faster R-CNN on real test datasets with motion blur.
In this paper we propose a novel approach to generate a synthetic aerial dataset for application in UAV monitoring. We propose to accentuate shape-based object representation by applying texture randomization. A diverse dataset with photorealism in all parameters such as shape, pose, lighting, scale, viewpoint, etc. except for atypical textures is created in a 3D modelling software Blender. Our approach specifically targets two conditions in aerial images where texture of objects is difficult to detect, namely challenging illumination and objects occupying only a small portion of the image. Experimental evaluation of YOLO and Faster R-CNN detectors trained on synthetic data with randomized textures confirmed our approach by increasing the mAP value (17 and 3.7 percentage points for YOLO; 20 and 1.1 percentage points for Faster R-CNN) on two test datasets of real images, both containing UAV-to-UAV images with motion blur. Testing on different domains, we conclude that the more the generalisation ability is put to the test, the more apparent are the advantages of the shape-based representation.