Aims and Scope:
The deep learning neural network has shown significant improvement in various fundamental computer vision tasks such as object detection, classification, recognition, segmentation, and tracking. The features extracted using deep learning are generic and expressive and could be used in many real-world applications. The UAV/drone-based aerial imagery has a significant role in emergency response and environmental applications, particularly in difficult and inaccessible areas. The use of drone-based imagery provides a perspective-distortion free view of the scene, which can significantly solve the impact of occlusion.

This special session aims to use drone-imagery for emergency response and damage assessment in disaster-stricken areas due to natural calamities such as floods, earthquakes, fire, and roadside accidents. It is vital to do the automated analysis of such events to provide a real-time alert for the emergency relief teams. Similarly, drone imagery is also widely used in the news and entertainment industry for surveillance and video recording of concerts and other gatherings. The potential applications might include counting and density applications, as well as the safety and security of participants. There are also significant potential applications in industry for the inspection of industries components such as pipeline and other confined spaces in the industry, infrastructure inspection and management, bridge health monitoring, to mention a few.

This special session aims to bring together the current research progress on Deep Learning architectures as well as applications for environmental monitoring, disaster management, crowd counting, industrial inspections, assessment of the damage caused by natural calamities, and various application in the entertainment industry.