| CARVIEW |
Who am I?
Coder, photographer, researcher, student, gamer. Interested in machine learning and computer vision.
Education
Resume
My current work and research focus is machine learning and computer vision. Below you can find my resume and LinkedIn.
Linkedin Download ResumeSelect Projects
Paper: Out-of-Distribution Detection Methods Answer the Wrong Questions
In this paper, we critically re-examine this popular family of OOD detection procedures, and we argue that these methods are fundamentally answering the wrong questions for OOD detection.
Arxiv
Modelverse
An online platform enabled by our algorithm for finding and sharing image generative models
Site Arxiv Youtube
Paper: Meta Deformation Network
A novel, dynamically generated neural network for faster inferencing and higher parameter efficiency for 3D human shape correspondence.
Arxiv
Audi-Exchange
AI-powered Hand Action Guidance to Assist Human-Human Interactions for the Blind and the Visually Impaired. Accepted to ACVR 2021 (https://iplab.dmi.unict.it/acvr2021/program) and presented as ICCV 2021 Workshop paper.
Paper Demo VideoPaper: Active Crowd Analysis for Pandemics
"Active Crowd Analysis for Pandemic Risk Mitigation for Blind or Visually Impaired Persons." Seondary Author, accepted to ACVR 2020 (https://iplab.dmi.unict.it/acvr2020/program)
Paper Supplementary
App: N&P Remote Mouse and Keyboard
Using your phone as your computer's keyboard and trackpad over Bluetooth or WiFi. A app I started in middle school with my good friend. 50K+ downloads.
Google PlayGithub: Air Guitar with Hand Detection
Play air guitar unironically with your fingers (kind of).
Github Demo Video
Github: Weakly Supervised Point-to-Tell
Learning with only classification labels (weak supervision) to segment images to determine the type of object someone is pointing at. Trained with custom a synthetic dataset that combines VOC and custom hand images.
GithubTech@NYU Talk: Intro to Computer Vision
A community talk about the basics of computer vision (neural networks,convolution operations, learning parameters, etc.).
Youtube