图灵讲坛(第34期):Learning to estimate 3D object pose from RGB images

发布时间:2019-05-06  阅读次数:291

 

活动时间:2019-5-10 13:30
活动地点:济事楼 455

 

Learning to estimate 3D object pose from RGB images

 

周晓巍

  浙江大学CAD&CG国家重点实验室研究员
  国家青年千人计划入选者

邀请人:贾金原 教授


Abstract:

Recovering the 3D pose of an object or human body relative to the camera is a fundamental problem in computer vision with many applications in robotics and augmented reality. Traditional methods attempted to solve this problem by aligning 3D models to 2D images with hand-crafted features, which lack robustness to appearance change and background clutter. Recently, deep learning has shown remarkable performances in feature learning. In this talk, I will summarize our recent efforts to combine geometric methods with deep learning for robust 3D pose estimation. First, I will introduce a pixel-wise voting network that robustly localizes keypoints and estimates 6DoF object pose even under severe occlusion. Next, I will discuss 3D human pose estimation and show that the lack of training data can be alleviated by using diverse supervision signals. Finally, I will present an approach for multi-person motion capture in a crowd scene with several calibrated cameras. 
 

Bio:

浙江大学CAD&CG国家重点实验室研究员,国家青年千人计划入选者。2008年本科毕业于浙江大学,2013年博士毕业于香港科技大学。2014年至2017年在美国宾夕法尼亚大学GRASP机器人实验室从事博士后研究。研究方向主要是计算机视觉及其在机器人、增强现实、医学影像分析等领域的应用,目前课题侧重于三维场景重建和语义计算,包括三维物体和人体的检测、识别、姿态估计、运动恢复、在线重建以及匹配等问题。策划和组织了Geometry Meets Deep Learning Workshops,并长期担任PAMI、IJCV、TIP等二十余种SCI期刊审稿人以及CVPR、ICCV、IJCAI等计算机领域顶级会程序委员会委员。
 

联系我们

地址:中国 上海曹安公路4800号同济大学软件学院

邮编:201804

联系电话:86-21-69589585,69589332(FAX)

 Copyright© 2017 同济大学软件学院

技术支持: 上海维程