Gengshan Yang | 杨庚山

I'm a MS/PhD (2017-) student at CMU Robotics, advised by Prof. Deva Ramanan. My research is focused on computer vision and learning.

I’m particularly interested in robust visual perception algorithms for mobile robots. At CMU, I'm working to develop depth and motion estimation algorithms that is 1) efficient, 2) generalizable and 3) able to produce uncertainty estimations.

I did my undergrad at Xi'an Jiaotong University, and worked with Prof. Chao Shen on motion sensors. In summer 2020, I did an internship in Google VisCAM. I was also fortunate to intern at Argo AI, TuSimple, and SVCL at UC San Diego.

Github /  Google Scholar  /  Email

A template-free approach for articulated shape reconstruction from a single video by combining differentiable rendering and data-driven correspondence and segmentation priors.

We propose a neural architecture powered by geometric reasoning that decomposes two frames into a rigid background and multiple moving rigid bodies, parameterized by 3D rigid transformations and depth.

We describe a neural architecture to upgrade 2D optical flow to 3D scene flow using optical expansion, which reveals changes in depth of scene elements over frames, e.g., things moving closer will get bigger.

We introduce several simple modifications to the optical flow volumetric layers that: 1) significantly reduces computation and parameters, 2) enables test-time adaptation of cost volume size, and 3) converges much faster.

To adress the problem of real-time stereo matching on high-res imagery, an end-to-end framework that searches for correspondences incrementally over a coarse-to-fine hierarchy is proposed.

We recast the continuous problem of depth regression as discrete binary classification, whose output is the occupancy probabilities on a 3D voxel grid. Such output reliably and efficiently captures multi-modal depth distributions in ambiguous cases.