About

In September 2019, I will be joining a new computer vision team at Apple (Seattle office) as an Applied Research Scientist. Our team is part of Apple AI and Machine Learning.

In my research, I focus on the intersection of geometric computer vision and machine learning, with the aim of developing efficient algorithms and techniques for accurate and real-time 3D reconstruction of the world. I recently graduated with a Ph.D. in Robotics from MIT, where I worked with Prof. Sertac Karaman and was affiliated with both LIDS and AeroAstro. Before coming to MIT, I obtained my bachelor's degree in computer engineering from HKUST in 2013.

Doctoral Thesis

Algorithms for Single-View Depth Image Estimation
Committee Members: Prof. Sertac Karaman, Prof. John Leonard, Prof. Nicholas Roy
[slides] [pdf]

Select Projects

Self-supervised Depth Completion
Fangchang Ma, Guilherme Venturelli Cavalheiro, Sertac Karaman.
Ranked 1st on the KITTI Depth Completion Competition in 2018
[ICRA'19] [code] [video] [bibtex]
Invertibility of Convolutional Generative Networks from Partial Measurements
Fangchang Ma, Ulas Ayaz, Sertac Karaman.
[NIPS'18] [bibtex] [supplementary] [code] [poster]
NVIDIA Pioneer Award
FastDepth: Fast Monocular Depth Estimation on Embedded Systems
Diana Wofk*, Fangchang Ma*, Tien-Ju Yang, Sertac Karaman, Vivienne Sze.
[ICRA'19] [code] [project webpage] [bibtex]
Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Image
Fangchang Ma, Sertac Karaman.
[ICRA'18] [code] [video] [bibtex]
Sparse Depth Sensing
Fangchang Ma, Luca Carlone, Ulas Ayaz, Sertac Karaman.
[IROS'16] [IJRR] [code] [video] [bibtex]
Lane Detection for Autonomous Tricycle
Persuasive Electric Vehicle (PEV)
On Sensing, Agility, and Computation Requirements for a Data-gathering Vehicle
Fangchang Ma, Sertac Karaman.
[WAFR'14] [extended version] [bibtex]
Velocity estimation from camera and IMU
Guyue Zhou, Fangchang Ma, Zexiang Li, Tao Wang .
[ROBIO'13] [bibtex]

Publications

ICRA'19
Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Camera
Fangchang Ma, Guilherme Venturelli Cavalheiro, Sertac Karaman. 2019 International Conference on Robotics and Automation. [pdf] [code] [video] [bibtex]
ICRA'19
FastDepth: Fast Monocular Depth Estimation on Embedded Systems
Diana Wofk*, Fangchang Ma*, Tien-Ju Yang, Sertac Karaman, Vivienne Sze. 2019 International Conference on Robotics and Automation.
[ICRA'19] [code] [project webpage] [bibtex]
NIPS'18
Invertibility of Convolutional Generative Networks from Partial Measurements
Fangchang Ma, Ulas Ayaz, Sertac Karaman. Conference on Neural Information Processing Systems. (NVIDIA Pioneer Award) [pdf] [bibtex] [supplementary] [code] [poster]
ICRA'18
Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Image
Fangchang Ma, Sertac Karaman. 2018 International Conference on Robotics and Automation.
[pdf] [code] [video] [bibtex]
IJRR
Sparse Depth Sensing for Resource-Constrained Robots
Fangchang Ma, Luca Carlone, Ulas Ayaz, Sertac Karaman. To appear in The International Journal of Robotics Research. [pdf] [code] [video] [bibtex]
IROS'16
Sparse Sensing for Resource-Constrained Depth Reconstruction
Fangchang Ma, Luca Carlone, Ulas Ayaz, Sertac Karaman. 2016 International Conference on Intelligent Robots and Systems. [pdf] [code] [video] [bibtex]
WAFR'14
Maximum-reward Motion in a Stochastic Environment: The Nonequilibrium Statistical Mechanics Perspective.
Fangchang Ma, Sertac Karaman. 2014 International Workshop on the Algorithmic Foundations of Robotics.
[pdf] [bibtex]
ROBIO'13
Velocity estimator via fusing inertial measurements and multiple feature correspondences from a single camera.
Guyue Zhou, Fangchang Ma, Zexiang Li, Tao Wang. 2013 International Conference on Robotics and Biomimetics.
[pdf] [bibtex]

Others