Projects / PennApps XXVI
Award winnerBest use of CVConfidence: low
Jarvis
Reach into the screen to grab, scale and rotate 3D models with your hands, and ask a voice assistant about them.
Project facts
- Team
- 4 people
- Build
- software only
- AI
- computer vision, voice ai
Problem
Spatial interaction usually needs a headset or special controllers.
Solution
A webcam tracks hands to pinch, spread and rotate meshes in a custom 3D pipeline, while a live voice model answers questions about the object on screen.
Demo moment
Team-described: pinch a model, spin it, and ask the assistant what it is.
Why it stood out
- organizer statedWon “Best use of CV” among 85 submissions.Source ↗
- team statedRecreates a famous film scene with a plain webcam.Source ↗
Lessons for your next hackathon
- Name the movie moment you are recreating; the judges fill in the rest.
Similar situations
Projects that share domains, mechanisms or judging lenses — often with a different problem and stack.Tune the comparison.
Award winner · 2025
Sculpt 3D scenes in the browser with your hands, assisted by AI.
Shared: domains creative media · mechanisms cultural reference hook, demo first wow · lenses wow factor · ai computer vision · stack flask, opencv
Podium #2 · 2026
Pick a place and a year on a 3D globe, and step into a generated 3D world of that moment with a voice guide.
Shared: domains education, creative media · mechanisms demo first wow · ai voice ai
Finalist · 2025
Voice-commanded robotic arm that sees its workspace and acts on natural-language instructions.
Shared: mechanisms cultural reference hook, demo first wow · lenses wow factor, technical depth · ai computer vision, voice ai · stack python, opencv, flask
Finalist · 2025
A robot that draws paintings onto a blind person's palm with warm and cold brushes, with narration.
Shared: domains creative media · mechanisms cultural reference hook, demo first wow · lenses wow factor, technical depth · ai computer vision, voice ai · stack python
Award winner · 2025
A first-person shooter controlled by a cardboard gun with a printed marker and a webcam.
Shared: mechanisms cost collapse · lenses technical depth, wow factor · ai computer vision · stack python, opencv, mediapipe
Finalist · 2025
Live 3D capture of people from three old Kinect sensors, merged into one hologram you can view in a headset.
Shared: mechanisms cultural reference hook, cost collapse, demo first wow · lenses wow factor, technical depth · stack opencv
Sources
- https://devpost.com/software/jarvis-wn0e48 · devpost project · retrieved 2026-10-13
- https://pennapps-xxvi.devpost.com/project-gallery · devpost gallery · retrieved 2026-10-13