Haitao Meng is currently a Ph.D. student in the Chair of Robotics, AI and Real-time Systems. His research advisor is Prof. Alois Knoll. He received his M.Eng degree in International Laboratory For Smart Systems at Northeastern University, China. Before he come to TUM to pursue his doctoral degree, he worked at both Sun Yat-sen University and Pengcheng Laboratory as a research assistant. His research interests include Stereo Estimation, Real-time Image Processing, and Neuromorphic Stereo Vision
Thesis Topic
Topic 1: VIsual-based Dense 3D Reconstruction
Stereo cameras are widely used in autonomous driving and robotics. It is cost-efficient and easy to deploy. The rich context information also enables the possibility of different task fusions, such as vision-based 3D object detection, segmentation, and tracking.
Topic 1.1: Real-time Pseudo-Lidar 3D Object Detection
Topic 1.2: Embedded-oriented Depth Prediction
Topic 2: Neuromorphic Stereo Vision
Inspired by biological retina, dynamical vision sensor transmits events of instantaneous changes in pixel intensity, giving it a series of advantages over the traditional frame-based cameras, such as high dynamical range, high temporal resolution and low power consumption. An interesting topic is to develop novel algorithms to predict depth with dynamical vision sensors to tackle extremely scenarios (e.g. low light, high dynamical scene, and high-speed objects)
Topic 2.1: Efficient Encodings and Representations of Event Stream Data
Topic 2.2: Learning to Reconstruct Dense Depth Map with Event Camera
If you are interested in one of the topics, please feel free to contact me.
Demo
This is a video demo accompanies the work: "FastFusion: Deep Stereo-LiDAR Fusion for Real-time High-precision Dense Depth Sensing"
Real-time 3D object datection based on stereo image pairs
Haitao Meng, Changcai Li, Gang Chen, Zonghua Gu and Alois Knoll: ER3D: An Efficient Real-time 3D Object Detection Framework for Autonomous Driving, 2023 mehr…
Meng, Haitao; Li, Changcai; Chen, Gang; Chen, Long; Knoll, ansd Alois: Efficient 3D Object Detection Based on Pseudo-LiDAR Representation. IEEE Transactions on Intelligent Vehicles, 2023, 1-12 mehr…
Meng, Haitao; Li, Changcai; Zhong, Chonghao; Gu, Jianfeng; Chen, Gang; Knoll, Alois: FastFusion: Deep stereo‐LiDAR fusion for real‐time high‐precision dense depth sensing. Journal of Field Robotics, 2023 mehr…
2021
Meng, Haitao; Zhong, Chonghao; Gu, Jianfeng; Chen, Gang: A GPU -accelerated Deep Stereo- LiDAR Fusion for Real-time High-precision Dense Depth Sensing. 2021 Design, Automation & Test in Europe Conference & Exhibition (DATE), IEEE, 2021 mehr…
2020
Chen, Guang and Wang, Haitao and Chen, Kai and Li, Zhijun and Song, Zida and Liu, Yinlong and Chen, Wenkai and Knoll, Alois: A survey of the four pillars for small object detection: Multiscale representation, contextual information, super-resolution, and region proposal. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2020 mehr…
