Xin Liu
Delft University of Technology
EDL P16-24 P6: Deep Learning for Mobile Robotics (MRob)
Research assignment: Deep Learning for Efficient Learning of Video Data |
Project Description: Efficient learning of video data is important for different video learning tasks, e.g. video object detection, video object segmentation, and tracking. Because it is infeasible to process each frame for video object detection. Applications require the utilization of temporal information to achieve higher detection performance. However, redundant frames in a video cost memory and speed. This project aims at developing efficient methods for learning video data, for example, save computation on redundant frames. |
Publications |
No frame left behind: Full Video Action Recognition. Xin Liu, Silvia L Pintea, Fatemeh Karimi Nejadasl, Olaf Booij, Jan C van Gemert. Conference proceedings of CVPR 2021. Open acces-gold. https://drive.google.com/open?id=1KZUneOkAYollikb4X_uij0AmwgxkQwzK |
| WeightAlign: Normalizing Activations by Weight Alignment. Xiangwei Shi, Yunqiang Li, Xin Liu, Jan van Gemert. Conference proceedings of ICPR 2020. Open access-gold. |
Personal information |
| LinkedIn Github |