Publications

The exploration of mutual-benefit cross-domains has shown great potential toward accurate self-supervised depth estimation. In this work, we revisit feature fusion ….

Recent visual object trackers provide strong dis- criminability towards accurate tracking under challenging sce- narios while neglecting the inference efficiency. Those methods handle all inputs …

This article describes the mechatronic design of a delivery octarotor drone equipped with safety features including protective bumpers for its propellers…

In this work, we explore the process of designing a long-term drone surveillance system by fusing object detection, tracking and classification methods…

A Pytorch based implementation for real-time object tracking on mobile devices. Optimization with Opencv, ONNX, etc. The speed achieves 46 Fps with single I7 core and performs state-of-the-art results

Recent object tracking methods depend upon deep networks or convoluted architectures. Most of those trackers can hardly meet real-time processing requirements on mobile platforms with limited computing resources. In this ….

In this work the visual tracking of an evading UAV using a pursuer-UAV is examined. The developed method combines principles of deep learning, optical flow, intra-frame homography ….

In this work, we propose a long-term UAV de- tection and tracking system from RGB-Thermal (RGB-T) sequences. The system consists of a high resolution daylight…

We develop a method to localize arbitrary-oriented text from unconstrained nature scene images. The method improved the automated information retrieval efficiency.