Xiaolong is a fourth-year PhD student in the Robotics Institute, Carnegie Mellon University, advised by Professor Abhinav Gupta. His research interests are computer vision and machine learning. He holds a MS in Computer Science from Sun Yat-sen University and BE in Software Engineering from South China Agricultural University.
Xiaolong’s research focuses on exploiting redundancy in visual data for learning visual representations. He has proposed to automatically extract supervisory signals from RGBD and video data where human labels are hard to come by. In these domains, his work utilizes the signals from depth and temporal information respectively to design auxiliary tasks to train convolutional neural network based RGB and RGBD representations. He also studies with modeling long-range pairwise relationships between redundant patterns in neural networks, which shows significant improvements on general video classification and image recognition tasks.