Skip to content
View WangHewei16's full-sized avatar
107 followers · 81 following
  • Carnegie Mellon University
  • Pittsburgh, United States
  • LinkedIn in/stephenw624

Achievements

Achievement: Galaxy BrainAchievement: QuickdrawAchievement: Pull SharkAchievement: Starstruck

Achievements

Achievement: Galaxy BrainAchievement: QuickdrawAchievement: Pull SharkAchievement: Starstruck

Block or report WangHewei16

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
WangHewei16/README.md

Hi! I am Stephen Hewei Wang👋, a master's student at Carnegie Mellon University, School of Computer Science (CMU SCS). I am working as Computer Vision Student Researcher at Meta (ex-Facebook) Reality Labs and Research Assistant at CMU Robotics Institute supervised by Dr. Ioannis Gkioulekas. I have previously interned at AppleVision Pro as Machine Learning Research Engineer. My interests lie in AR/VR, LLMs, VLM, computer vision, machine learning, and foundation models. I graduated from University College Dublin (UCD) in Ireland, earning a Bachelor’s in Software Engineering with Cum Laude. During UCD, I took on a role as an ML/CV research intern in THEIA lab and collaborated with Nanyang Technological University (NTU), supervised by Assoc Prof Yee Hui Lee and Dr. Soumyabrata Dev. My endeavors have culminated in 10+ works with 150+ citations published and delivered in esteemed conferences, journals, and workshops including CVPR, AAAI, CIKM, BMVC, IEEE, ACM, SCI, and Elsevier.

🔍 Linkedin | Google Scholar | ResearchGate

Github stats

Pinned Loading

  1. DMCNet-for-Video-Engagement-Understanding DMCNet-for-Video-Engagement-Understanding Public

    [Elsevier SOCL2022] Investigate in ML/DL-ensembled models and visualize features by dimension reduction techniques like PCA and t-SNE, measure performance via multiple metrics (e.g., Gini Index, AG…

    Jupyter Notebook 25 2

  2. AMDCNet-for-Stereo-Matching AMDCNet-for-Stereo-Matching Public

    [Elsevier Displays 2022] Propose visual sensitivity regularization in cost calculation stage and multi-directional aggregation template, optimize parallax through left-right consistency detection.

    C++ 24 6

  3. 3D-VistaNet-Purificatory-LoFTR-and-DKM-with-Test-Time-Augmentation-for-Scalable-Image-Matching 3D-VistaNet-Purificatory-LoFTR-and-DKM-with-Test-Time-Augmentation-for-Scalable-Image-Matching Public

    [CVPR2022 Image Matching Workshop (IMW) Top6%] Propose a model registering two images from diverse viewpoints with coarse-to-fine attention, use LoFTR for local feature matching and DKM for regress…

    Jupyter Notebook 20 3

  4. CassavaNet-An-EfficientNet-Vision-Transforer-for-Real-Time-Leaf-Disease-Identification CassavaNet-An-EfficientNet-Vision-Transforer-for-Real-Time-Leaf-Disease-Identification Public

    [Kaggle2020 Competition Top1.59% Silver Medal🥈] Develop an EfficientNet-based DL network consisting MBConv modules and a ViT backbone to classify the disease types of cassava leaf.

    Jupyter Notebook 10 2

  5. MelodyDreambit-AI-driven-Musical-Instrument-E-commerce-Website MelodyDreambit-AI-driven-Musical-Instrument-E-commerce-Website Public

    [Website based on Flask/SpringBoot+Vue2] Develope a multi-lingual trade website with customer and staff portals; enabled end-to-end functions such as online chatting, order placement, shipment trac…

    JavaScript 1

  6. Stroke-Prediction-from-Electronic-Health-Records Stroke-Prediction-from-Electronic-Health-Records Public

    Forked from Soumyabrata/EHR-features

    [Elsevier Healthcare Analytics 2022] Analyze vital factors in EHR to predict stroke, perform correlation analysis and adopt PCA to identify patterns in low-dimension subspace of the feature space.

    PostScript 8