Hao Liang

About

I am an ECE PhD student at Rice University advised by Professor Guha Balakrishnan. Before that, I spent 3 wonderful years at Carnegie Mellon University, where I was fortunate to work closely with Professor Gauri Joshi and Professor Bhiksha Raj. I did my undergrad at UESTC. My research focuses on generative modeling, 3D human reconstruction, vision–language models and Federated Learning. I like building practical systems that are simple, fast, fair and reliable. My current research is supported by Samsung Research America.

Publications

1. FastAvatar: Instant 3D Gaussian Splatting for Faces from Single Unconstrained Poses
   Hao Liang, Zhixuan Ge, Ashish Tiwari, Soumendu Majee, G.M. Dilshan Godaliyadda, Ashok Veeraraghavan, Guha Balakrishnan. In submission
   PDF · Code · Project page
2. Linking convolutional kernel size to generalization bias in face analysis CNNs
   Hao Liang, Josue Ortega Caro, Vikram Maheshri, Ankit B Patel, Guha Balakrishnan. WACV '24
   PDF
3. Benchmarking Algorithmic Bias in Face Recognition: An Experimental Approach Using Synthetic Faces and Human Evaluation
   Hao Liang, Pietro Perona, Guha Balakrishnan. ICCV '23
   PDF · Project page · Code · Dataset
4. X-ray Recognition: Patient identification from X-rays using a contrastive objective
   Hao Liang, Kevin Ni, Guha Balakrishnan. arXiv '24
   PDF
5. Visualizing chest X-ray dataset biases using GANs
   Hao Liang, Kevin Ni, Guha Balakrishnan. MIDL '23
   PDF
6. RareGAN: Generating Samples for Rare Classes
   Zinan Lin, Hao Liang, Giulia Fanti, Vyas Sekar. AAAI '22
   PDF · Code
7. A Novel Framework for the Analysis and Design of Heterogeneous Federated Learning
   Jianyu Wang, Qinghua Liu, Hao Liang, Gauri Joshi, H Vincent Poor. IEEE Transactions on Signal Processing
   PDF
8. Controlled AutoEncoders to Generate Faces from Voices
   Hao Liang, Lulan Yu, Guikang Xu, Bhiksha Raj, Rita Singh. ISVC '20
   PDF
9. Tackling the objective inconsistency problem in heterogeneous federated optimization
   Jianyu Wang, Qinghua Liu, Hao Liang, Gauri Joshi, H Vincent Poor. NeurIPS '20
   PDF · Code
10. Overlap local-SGD: An algorithmic approach to hide communication delays in distributed SGD
    Jianyu Wang, Hao Liang, Gauri Joshi. ICASSP '20
    PDF

Teaching

• ELEC 566: Introduction to Computer Vision, TA, Rice University · Fall 2024.
• ELEC 542: Neural Methods for Image Synthesis, TA, Rice University · Spring 2022.
• 11-785: Introduction to Deep Learning, TA, Carnegie Mellon University · Fall 2019.

Let's Connect

Feel free to reach out for collaboration opportunities, research discussions, or just to say hello.

Houston, TX