Boxiao (Leo) Pan 🦁

Master Student at Stanford University
bxpan [at] stanford [dot] edu / leobxpan [at] gmail [dot] com
LinkedIn / GitHub / Twitter


Hi! I'm a Master student at Stanford Vision & Learning Lab, where I've been fortunate working with Prof. Juan Carlos Niebles and Prof. Fei-Fei Li. I'm fascinated by building intelligent systems that can interpret and understand human-centered actions, scenes, and events, especially from video input. Before coming to Stanford, I obtained my B.E. degree from South China University of Technology.

News

  • [new] 2020.02 One paper accepted to CVPR 2020.

  • 2019.11 Our paper on Cross-Domain Action Recognition is accepted to AAAI 2020 as an oral presentation! See you in New York! 🗽

  • 2019.10 I'm applying to Ph.D. programs for Fall 2020.


Publications

Spatio-Temporal Graph for Video Captioning with Knowledge Distillation
Boxiao Pan, Haoye Cai, De-An Huang, Kuan-Hui Lee, Adrien Gaidon, Ehsan Adeli, Juan Carlos Niebles
Conference on Computer Vision and Pattern Recognition (CVPR), 2020.
Paper

Adversarial Cross-Domain Action Recognition with Co-Attention
Boxiao Pan*, Zhangjie Cao*, Ehsan Adeli, Juan Carlos Niebles
AAAI Conference on Artificial Intelligence (AAAI), 2020. (Oral Presentation)
Paper Slides Poster

A Novel LSTM-Based Daily Airline Demand Forecasting Method Using Vertical And Horizontal Time Series
Boxiao Pan, Dongfeng Yuan, Weiwei Sun, Cong Liang, Dongyang Li
Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD) Workshop, 2018.
Paper

Teaching

AI4ALL: During Summer 2019, I was a research mentor for the Computer Vision team at Stanford AI4ALL, along with Andrew Kondrich. We spent three amazing weeks with eight talented high-school girls working together on a project, which applied Computer Vision techniques on Global Poverty Discovery. Aside from the project, we also designed course materials, assignments and gave lectures.

Graduate Teaching Assistant: At Stanford, I've been serving as a TA for CS131: Computer Vision: Foundations and Applications (Fall 2019) and CS231n: Convolutional Neural Networks for Visual Recognition (Spring 2019).

Projects

Self-supervised Motion Retargeting with Single Frame
CS448v: Computational Video Manipulation
with Karttikeya Mangalam.
Report

Tackling Occlusion in Person Re-identification
CS230: Deep Learning
with Soutik Chakraborty and Pragya Mishra.
Report Poster

Question Answering on SQuAD2.0
CS224n: Natural Language Processing with Deep Learning
with Gael Colas and Shervine Amidi.
Report Poster

Predicting NYC Taxi Fares, Trip Distance, and Activity
CS229: Machine Learning
with Paul Jolly and Varun Nambiar.
Report Poster

Lollipop: Mobile App for South China University of Technology
with 100steps Student Association.

Misc

  • I mostly enjoy music as my secondary life. Fun fact: I've been playing piano since 4 :P

  • I love traveling. I maintain a Places I've Visited map. Looking forward to exploring the rest of the world!