Estimating poses of three dimensional (3D) objects is of great importance to many high level tasks such as robotic manipulation, scene interpretation and augmented reality. Detecting poorly textured objects and estimating their 3D pose is still a challenging problem. The objective and expected result of this research is to develop a systematic and applicable approach that could detect poorly textured 3D object pose. The proposed method is using state-of-the art deep learning in computer vision. The proposed research is related to the interns PhD research and is an ongoing research for the partner organization. The collaboration could tie industry expert and academia, and turn many concepts to commercialization, and bring ideas/theories to life and real applications.
Clarence de Silva
Vancouver Computer Vision Ltd
Engineering - mechanical
Information and communications technologies
Find the perfect opportunity to put your academic skills and knowledge into practice!Find Projects