Jiajun Wu
Jiajun Wu is an Assistant Professor of Computer Science at Stanford University. Research. My group studies physical scene understanding---building machines that see, reason about, and …
Jiajun Wu Stanford University [email protected] Abstract Grounding object properties and relations in 3D scenes is a prerequisite for a wide range of artificial intelligence tasks, such as …
There have been some works that aim to estimate physical object properties [Wu et al.,2016,2015, Denil et al.,2017].Wu et al.[2015] explored an analysis-by-synthesis approach …
Jiajun Wu† Stanford University Abstract We introduce Programmatic Motion Concepts, a hierar-chical motion representation for human actions that cap-tures both low-level motion and high …
de-animation (VDA) (Wu, Lu, et al. 2017), our model learns to jointly infer physical world states and simu- late scene dynamics, integrating both a physics engine
reconstruction [Wu et al.,2016a,Yan et al.,2016,Rezende et al.,2016,Tulsiani et al.,2017]. In our paper, we exploit this idea to develop differentiable constraints on the consistency between …
JIAJUN WU, Stanford University BO ZHU, Georgia Institute of Technology, Dartmouth College Fig. 1. Inkdrop simulated by Neural Flow Maps, a novel, high-fidelity simulation system built …
Jiajun Wu MIT CSAIL [email protected] ABSTRACT We propose the Neuro-Symbolic Concept Learner (NS-CL), a model that learns visual concepts, words, and semantic parsing of …
Shangzhe Wu University of Oxford Noah Snavely Cornell Tech Jiajun Wu Stanford University One Training Image Object Intrinsics e o Ò5000Ó Generated Roses Figure 1. From a single image, …
Jiajun Wu3 Joshua B. Tenenbaum 4Antonio Torralba Abstract Humans intuitively recognize objects’ physical properties and predict their motion, even when the objects are engaged in …