Tan M. Dinh

I am an AI Engineer at VinAI. Before that, I was an AI Resident at here, where I conducted research on computer vision and deep generative models, under the supervision of Dr. Binh-Son Hua, Dr. Rang Nguyen, and Dr. Anh Tran.

I obtained my Bachelor of Engineering Degree on Computer Science in Honors Program with the highest rank at Ho Chi Minh City University of Technology (HCMUT) in 2020.

I am interested in computer vision, GAN-inversion and applications such as real face reconstruction and manipulation, language and vision learning such as text-to-image synthesis.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

profile photo
Publications
HyperInverter: Improving StyleGAN Inversion via Hypernetwork
Tan M. Dinh, Anh Tran, Rang Nguyen, Binh-Son Hua
CVPR 2022
paper / project page / code

We develop a novel entirely encoder-based two-stage StyleGAN inversion method, which can obtain the superiority of reconstruction quality, editability, and fast inference at the same time by employing the idea of hypernetworks.

TISE: Bag of Metrics for Text-to-Image Synthesis Evaluation
Tan M. Dinh, Rang Nguyen, Binh-Son Hua
ECCV 2022
paper / code

We identify several pitfalls in the existing evaluation pipeline of text-to-image synthesis models, and propose a set of metrics combining the improved versions of existing metrics and new ones to form a unified evaluation toolbox, so-called TISE, for benchmarking text-to-images models fairly, robustly, and consistently.

Work Experience

  • [February 2022 - present]    AI Engineer @ VinAI
  • [April 2020 - January 2022]   AI Research Resident, Computer Vision Group @ VinAI
  • [April 2019 - January 2020]   Data Scientist Collaborator, R&D Lab @ Zalo, VNG

Awards & Honors

Professional Services
Reviewer: CVPR 2022, ECCV 2022

Last update: April 9, 2022


▶ di-mi-ta.github.io's clustrmaps 🌎

Website template by Jon Barron