Abstract:The target tracking algorithm is suffering from low accuracy and poor applicability due to the complex lighting conditions and severe changes in scale caused by the relative lens movement during panoramic video target tracking. To address this issue, we propose an algorithm for panoramic target tracking based on the improved SiameseRPN. First, the network structure of MobileNetV3 is used to extract the deep features to make the algorithm have a better adaptability to the scene defects appearing in panoramic video sequences, and the Squeeze and Excite module is used to increase the sensitivity of the network to feature selection. Then, we propose and construct a feature fusion module based on bilinear interpolation, which is used to make the output depth features of the last three layers have the same scale, and these three layers of features arc fused for network prediction. Finally, we use a classification sequence to predict the positive and negative samples in the current sequence, and adopt a regression branch to predict the position information and scale information of current output targets. Thus the target position information is outputted. The experimental results show that the proposed algorithm has better tracking accuracy and it can effectively deal with the problems of poor local image quality and scale changes in panoramic data, while maintaining the real-time tracking performance. It shows a good adaptability to small targets, target occlusion, and multi-target cross movements in target tracking, and has good visual effects and high tracking scores.

Robot Target Recognition and Tracking Based on Panoramic Vision

Algorithm for Panoramic Video Tracking Based on Improved SiameseRPN

Multi-features Guided Robust Visual Tracking.

A Real-time Target Detection Algorithm for Panorama Infrared Search and Track System

Research on target tracking of binocular vision robot

An Autonomous Target-Tracking Algorithm Based on Visual Feature

Dynamic Object Recognition and Tracking of Mobile Robot by Monocular Vision

A robust target recognition and tracking panoramic surveillance system based on deep learning

Research on Robot Vision Target Tracking Detection Method Based on Image Segmentation.

Panoramic Multi-Target Tracking Model Incorporating Human Features and Motion Model

Target-tracking Algorithm for Omnidirectional Vision.

Small Surface Target Detection Algorithm Based on Panoramic Vision

Object Detection Algorithm Based on Panoramic Stereo Video

Target Recognition and Localization of Mobile Robot with Monocular PTZ Camera

Vision Based Target Tracking Algorithm in Natural Complex Scene

Ptz Camera Target Tracking in Large Complex Scenes

Real-time Object Tracking of a Robot Head Based on Multiple Visual Cues Integration.

Real-time Pose Tracking of 3D Targets by Photo-Model-based Stereo-Vision

People Tracking in Panoramic Video for Guiding Robots

Research on high resolution & high sensitivity panoramic surveillance system

Target Tracking and Measuring Based on Binocular Vision