Primary supervisorJianfei Cai
Deep learning has achieved ground-breaking performance in many 2D vision tasks in the recent years. With more and more 3D data available such as those captured by Lidar, the next research trend is doing advanced perception on 3D data. The objective of this project is to study the state-of-the-art object detection techniques for 3D point clouds such as PointNet and PointVoxel.
This is a "research project" best for students who are independent and willing to take up challenges with high expectation in the grade when fulfilled the somewhat challenging requirements. Under-performing is likely to fail to obtain the passing requirements. It is also a good practice for students who wish to pursue further study at a postgraduate/PhD level.
The objective of this project is to study the state-of-the-art deep learning based object detection techniques for 3D point clouds such as PointNet and PointVoxel. It can be applied in driverless cars to fast and accurately detect objects such as traffic cones, people and lanes.
- PointNet++: https://arxiv.org/abs/1706.02413
- PointVoxel CNN: https://arxiv.org/abs/1907.03739
The student must have knowledge on deep learning (e.g. taking online Stanford deep learning, computer vision related courses) and is skillful in Python programming.