24小时热门版块排行榜

返回列表

HiChen.

新虫 (初入文坛)

应助: 0 (幼儿园)
金币: 70
帖子: 33
在线: 4.8小时
虫号: 28420110
注册: 2022-02-21
专业: 力学

[交流] 西交利物浦大学机器视觉及人工智能研究方向项目招聘

动态交通感知的神经符号技术研究
Neuron-Symbolic Technology of Dynamic Transportation Perception
培养方式：西交利物浦大学 + 集萃深度感知所联合培养
培养地点：无锡
博士导师：Prof. Steven Guan (西交利物浦大学); Dr. Yutao Yue (集萃深度感知所); Prof. Eng Gee Lim (西交利物浦大学); Prof. Prudence Wong (英国利物浦大学)。采用产教研深度融合的联合导师机制。
博士学位授予单位：英国利物浦大学
培养资助：免学费奖学金（抵扣学费折合80000元/年）；博士生在集萃深度感知所联合培养期间，联合培养单位提供3000-6000元/月生活补助。

Project Description
Cameras can perceive the appearance, color and shape information of objects in the environment, and are widely used in target classification, target detection, target segmentation, target tracking and other fields. With the development of deep learning, although significant progress has been made in general target detection, the detection of vehicles and pedestrians in autonomous driving is different from general target detection, so there are still many challenges in the detection of vehicles and pedestrians in the driving environment. Based on the video or image captured by road conditions, the scale distribution range of vehicles and pedestrians is very wide, and small and medium-sized targets account for a large proportion. The existing target detection algorithms have low accuracy in detecting small and medium-sized vehicles and pedestrian targets. It is necessary to study the multi-scale vehicle and pedestrian detection under real road conditions, and study the detection algorithm that is suitable for multi-scale targets and can significantly improve the detection effect of small and medium-scale targets.
摄像头可以感知所处环境中物体的外貌、颜色和形状信息，被广泛应用在目标分类、目标检测、目标分割、目标跟踪等领域。随着深度学习的发展，通用目标检测虽然取得了显著的进展，但是由于自动驾驶中车辆和行人的检测与通用目标检测不同，所以在驾驶环境下的车辆和行人检测仍旧存在很多挑战.。基于路况拍摄的视频或者图像，车辆和行人的尺度分布范围很广，中小尺度目标占比很大，现有的目标检测算法对中小尺度的车辆和行人目标检测精度很低。需要针对真实路况下多尺度的车辆与行人检测进行了研究，研究适用多尺度目标，对中小尺度目标检测效果有显著提升的检测算法。

投递简历或项目课题相关问题请联系导师：
Prof. Steven Guan（西交利物浦大学，Principal Supervisor）
Email: steven.guan@xjtlu.edu.cn
Dr. Yutao Yue（集萃深度感知所，JITRI Supervisor）
Email: yueyutao@idpt.org

博士申请相关问题请咨询研究生院：
邮箱：DoctoralStudies@xjtlu.edu.cn
电话：0512-81889001转3

入学要求：
学术要求：
申请者需为数学、物理、计算机、自动化等相关专业背景，有较强的机器学习领域知识背景。
英语语言成绩要求：
雅思6.5分（单项不低于5.5分）或同等水平托福、PTE成绩。

申请材料：
博士研究课题申请书
中、英文完整成绩单
中、英文学位证明
学信网学历认证/境外学历认证
英语语言成绩证明
英文个人陈述
两封推荐信

回复此楼