[May 5, 2017] I am serving the program chair of 3DV 2017.
[Mar 31, 2017] I am organizing the IEEE Workshop on Learning to See from 3D Data in conjunction with ICCV 2017, Venice, Italy.
[Mar 31, 2017] I have one paper accepted at SIGGRAPH 2017.
[Feb 27, 2017] I have 5 papers (2 first author orals, 1 spotlight, 2 posters) accepted at CVPR 2017.
[Feb 24, 2017] I am giving an invited talk at UC San Diego in April.
[Feb 10, 2017] I am giving an invited talk at Cornell University in April.
[Jan 17, 2017] I am organzing a tutorial on 3D Deep Learning at CVPR 2017, Hawaii, USA.
[Dec 15, 2016] Posted the slides of my recent talks on 3D representation learning and synthesis for learning. They summarize the majority of my efforts in the past 3 years.
[Nov 24, 2016] I am giving talks at MIT (Brain and Cognitive Sciences Department and CSAIL), on 3D object reconstruction and abstraction by deep learning.
I will give a talk at Adobe and Facebook on understanding geometries from sensor data.
I have published 1 paper at NIPS, 3 papers at SIGGRAPH Asia, 1 paper at ECCV (spotlight), 2 papers at CVPR (1 spotlight) in 2016, and 1 paper at 3DV (oral) in 2016.
To visit Europe for two weeks in early November of 2015 and give talks on Joint Analysis for 2D Images and 3D shapes at several institutes, including Max Planck Institute (Germany), Ecole Polytechnique (France), Ecole Normale SupA(C)rieure (France), Tel Aviv University (Israel) and The Hebrew University of Jerusalem (Israel).
Talk on Joint Analysis for 2D Images and 3D shapes at UC Berkeley, Oct 16, 2015.
I have two oral papers accepted by ICCV 2015 (2% acceptance rate for oral presentation).
The mission and big picture of research happening in SU Lab --- learning to interact with the environment.
It describes the extension of SU Lab's research focus from deep 3D representation learning to broader topics of artifical intelligence for interacting with the environment. Not all papers published in the year are included in the report. Missing topics are binary neural networks and adversarial defense.
Towards Attack-Agnostic Defense for 2D and 3D Recognition
Invited talk at the Workshop of AdvML in CVPR2019 (updated on July 3, 2019)
A summary of the work on 2D/3D adversarial defense in 2018-2019. The main messages are: (1) Lower-dimensional data seems to be easier to defend; and (2)Defending in lower resolution seems to be more attack agnostic.
Synthesize for Learning
Invited talk at 3DV workshop on Understanding 3D and Visuo-Motor Learning (updated in Sep, 2016)
Use synthetic data to train learning algorithms for applications such as viewpoint estimation, human pose estimation, and robot perception. Based upon 5 recent papers of mine.
