This talk tells the story of virtual unwrapping, conceived during the rise of digital libraries, computer vision, and large-scale computing, and now realized on some of the most difficult and iconic material in the world - the Herculaneum Scrolls - as a result of the recent phenomena of big data and machine learning. Virtual unwrapping is a non-invasive restoration pathway for damaged written material, allowing texts to be read from objects that are too damaged even to be opened. The Herculaneum papyrus scrolls, buried and carbonized by the eruption of Mount Vesuvius in 79 CE and then excavated in the 18th century, are original, classical texts from the shelves of the only library to have survived from antiquity. The 250-year history of science and technology applied to the challenge of opening and then reading them has created a fragmentary, damaged window into their literary and philosophical secrets. In 1999, with more than 400 scrolls still unopened, methods for physical unwrapping were permanently halted. The intact scrolls present an enigmatic challenge: preserved by the fury of Vesuvius, yet still lost. Using a non-invasive imaging approach, we have now shown how to recover their texts, rendering them "unlost." The path we have forged uses high energy physics, artificial intelligence, and the collective power of a global, scientific community inspired by prizes, collaborative generosity, and the common goal of shared glory: reading original classical texts for the first time in 2000 years.
- Layered Diffusion Brushes, Peyman Gholami, Robert Xiao
- Semantic navigation, Yue Hu
- Demo for One-Step Specular Highlight Removal with Adapted Diffusion Models, Mahir Atmis, Levent Karacan, Mehmet Sarıgul
- Image as an IMU, Jerred Chen, Ronald Clark
- Underwater NeRFs with an AUV, Selena Sun, Elsa McElhinney, Vassilis Alexopoulous, Felicie Hoffman, Lawton Scaling, Kai Song, Rohan Bhowmik, Angelina Krinos, Ryota Sato, Chisa Ogaki, Julian Shultz, Professor Oussama Khatib
- Scalable Object Detection in Mixed Reality using Incremental Re- training and One-shot 3D Annotation, Alireza Taheritajar
- SOAP: Style-Omniscient Animatable Portraits, Tingting Liao
- Vision-guided Shared Autonomy for Smarter Prosthetics Federico Vasile