Pramish Paudel

At INSAIT (Sofia), I work with Prof. Luc Van Gool, Dr. Danda Pani Paudel, and Dr. Ajad Chhatkuli to build a Jarvis-like XR copilot for aircraft maintenance—an assistant that sees what a technician sees, understands tools/parts and context, verifies each SOP step in real time, and leaves audit-ready evidence.

Currently: Building an AR verification co-pilot for aircraft maintenance that watches each step against standard operating procedures in real-time and produces tamper-evident audit trails; piloting integrations with MRO systems (e.g., AMOS) and head-worn devices.

My long-term goal: To create AI models (accurate representation of our complex, 3D and dynamic world) that can teach any concept to any person through any modality, thereby making learning anything universally accessible and transformative.

Email: pramish[dot]paudel[at]gmail[dot]com  /  Scholar  /  Github

profile photo

Publications

iHuman: Instant Animatable Digital Humans From Monocular Videos
Pramish Paudel, Anubhav Khanal, Danda Pani Paudel, Jyoti Tandukar, Ajad Chhatkuli
ECCV, 2024
project page / video / arXiv

iHuman lets you reconstruct high fidelity 3D Gaussian Splats and 3D Mesh from monocular video of moving humans.

Awards and Honors

  • Summer Undergraduate Research Fellowship, INSAIT: Selected in INSAIT's prestigious Summer Research Fellowship for undergraduate students all over the world with acceptance rate as low as 3%.
  • Hult Prize Global Accelerator 2022, Boston, US: Reached to top 16 teams worldwide competing over 30K teams with my startup MeasureMe.AI.
  • Huawei Seeds for the Future 2022: Recognized as one of the top 7 computer science talents nationwide.
  • Start-up Seed Fund Grant Recipient, 2023: Secured NPR 3.2 Million as a startup running fund for my college department through my startup.
  • Microsoft Imagine Cup 2022: Represented South Asia at the Asia Regionals.
  • ICT Award 2023: Finalist among Nepal’s top twelve innovators.
  • Hack A Week 2018, 2019: First place in AI-themed hackathon at LOCUS.
  • KU IT MEET 2019: Winner of the hackathon hosted by Kathmandu University.
  • Glocals’ 20 Under 20, 2017: Honored for driving notable societal change through IT.

Projects

  • [2019] Radiograph Abnormality Detection (RAD): Developed an X-ray image abnormality detection system that excelled in competitions, winning first place at KU IT MEET 2019 and Locus 2019 AI hackathons.
  • [2018] VAE GAN Anime Image Colorizer: Created an image colorization tool that won the Locus 2018 national level hackathon while still in high school.
  • [2021] AutoDiff Package: Built a C++ package for automatic differentiation using dual numbers.
  • [2022] Blockchain from Scratch: Developed a blockchain system focusing on proof of work algorithms, enhancing understanding of blockchain fundamentals.
  • [2020] Vicaaya - Movie Search Engine: Developed a movie search engine using Elasticsearch for search capabilities. Engineered Scrapy scraper to scrape 100M+ items throughout the internet with very coordinated distributed instances.
  • [2017] Carduino: Designed a mini self-driving car model utilizing smartphone camera and ultrasonic sensors, awarded at the Trinity SciTech Expo.
  • [2016] Proto News / Khabar Sanjal: Created a news app for Nepalese abroad at age 15, reaching over 20,000 downloads.

Blogs

Books


The template is borrowed from Jon Barron