News 🚀

  • 📄 New Paper: KnobGen has been accepted to CVPR 2025 workshop proceedings!
  • 💼 New Role: Started computer vision research intern role at Path Robotics Inc!
  • 📄 New Paper: SegFormer3D has been accepted to CVPR 2024 workshop proceedings!
  • 🏆 New Award: Won Robert E. Alternhofen Memorial Scholarship award!

    • Publications 📚

    KnobGen: Controlling The Sophistication of Artwork in Sketch-Based Diffusion Models

    KnobGen
    📌 Novice or professional sketch? We've got you in both🎨
    KnobGen offers a flexible conditional sketch guided image diffusion model that caters to broad range of sketch complexities from amateurish to professional art.
    CVPR, 2025.
    PDF CODE

    SegFormer3D: an Efficient Transformer for 3D Medical Image Segmentation

    SegFormer3D
    📌 Limited compute? No Problem!
    Segformer3D is a light-weight and efficient hierarchical Transformer designed for 3D volumetric segmentation. It is on-par with large-scale models such as nnFormer, Unetr and SwinUnetr. SegFormer3D leverages from an all-MLP decoder to aggregate local and global attention features to produce highly accurate segmentation masks. We benchmark our model on BraTs, Synapse and ACDC.
    CVPR, 2024.
    PDF CODE

    A Probabilistic-based Drift Correction Module for Visual Inertial SLAMs

    📌 when visual-inertial navigation system becomes inaccurate!
    This paper presents a probabilistic-based drift correction module for improving the accuracy of Visual Inertial Simultaneous Localization and Mapping (SLAM) systems. The proposed method treats positioning measurements as stochastic variables in a multivariate distribution, incorporating geospatial priors to correct accumulated drift error.
    ISPRS, 2024.
    PDF

    Experience 🔎

  • Computer Vision Researcher Intern @ Path Robotics Inc., Nov 2024 ‑ Present, Columbus, Ohio, US
    • Photorealistic conditional (lighting and material) image generation of 3D objects using 3D diffusion models
    • World Models for autonomous welding robots
    • Active learning on out‑of‑distribution samples
    • Large‑scale pipeline refactoring for optimized ML experimentation
    3D diffusion demo
  • Graduate Research Associate @ Photogrammetric Computer Vision Lab, Feb 2021 ‑ Aug 2025 (expected), Columbus, Ohio, US
    • 3D video understanding and classification
    • Self-supervised learning on large-scale 3D video
    • Biomimetic 3D (bio) medical image understanding
    • 3D Ultrasound (ocular and echocardiogram) Sequence Understanding
  • Simultaneous Localization and Mapping Team Lead @ The General Motors SAE Autodrive Challenge, Oct 2022‑ May 2023, Ohio, Columbus, US
    • Accurate SLAM in narrow angle field of view scenarios

    • Services 🔧

  • Reviewer for CVPR, ECCV, ICCV, ICLR, AVSS, ACCV, SIBGRAPI conferences, 2023-2025
  • Large-scale distributed data parallel and training @ PCVLab, 2024-2025

    • Contact 📧