Projectibles: Optimizing Surface Color For Projection
summary
Combines an off‑the‑shelf printer and projector with two decomposition algorithms that split each video frame into a printed component and a projected component to maximize contrast and resolution while avoiding projector clipping. Measured results: printer ≈ 34:1, projector ≈ 200:1 (drops to 29:1 with ambient) and combined ≈ 700:1; supports up to 35 MP stitched footage, MATLAB optimization takes about 3.5 hours for a 480px image but a heuristic with artist controls runs in ~0.5 s, six sequences and a 10‑user study show higher satisfaction and no perceptible artifacts.
AI-written summary · about
chapters (8)
- 0:00Introduction to Projectibles
- 1:51Combining static and dynamic displays
- 3:17Hardware and contrast improvements
- 4:18Real-world examples and results
- 6:47Mathematical model and optimization
- 11:44Heuristic method and artist workflow
- 12:46Validation and user study
- 13:28Q&A session
related talks
03mJust realized we had a “Time Machine” after Demuxed’18 day 1
03mDelivering fast and beautiful images and video
31mDeveloping VR Media Technology
09mHow to Throw Out 95% of Pixels in VR, Without Anyone Noticing!!
10mVideo vectorization
25mOptimal Video Compression Using Pixel Shift Tracking
19mRethinking the Video Transcoding Pipeline: How Pre-Filtering Holds the Key to Efficiency and Quality
32mMaking of this Stage by { live : js }