Max Robotics — Global Service Robot Directory + US Certification

About

Chuhang Zou is a computer vision researcher whose work centers on 3D scene understanding, shape representation, and the recovery of structured geometric information from limited sensor data. His most influential contribution, "3D-PRNN: Generating Shape Primitives with Recurrent Neural Networks" (2017), has garnered nearly 200 citations and introduced a novel framework for representing complex 3D shapes as collections of simple geometric primitives using recurrent neural networks — a approach directly inspired by how humans perceive and decompose objects in the physical world. This work has broad implications for robotics, digital content creation, and 3D visualization. Zou has also made significant strides in indoor scene parsing, with his 2015 paper "Predicting Complete 3D Models of Indoor Scenes" (51 citations) demonstrating how full scene layouts — including occluded surfaces and objects — can be inferred from a single RGBD image. His follow-up work on complete 3D scene parsing further refined these methods by modeling objects as detailed CAD models and recovering both visible and hidden scene geometry. Collectively, Zou's research advances the field's ability to turn raw sensor input into rich, actionable 3D representations of the real world.