Scholar
Charlie Hewitt
Google Scholar ID: MkqRg64AAAAJ
Google
Citations & Impact
Citations
1,127
H-index
10
i10-index
10
Publications
20
Co-authors
5
list available
Publications
Resume (English only)
Academic Achievements
- - VoluMe: Real-time 3D Gaussian reconstruction technology
- - DAViD: Application of small-scale, high-fidelity synthetic datasets in computer vision tasks
- - GASP: Training Gaussian Avatars with Synthetic Priors
- - Look Ma, no markers!: Marker-free, high-quality full-body reconstruction technique
- - Hairmony: Method for predicting hairstyles from a single image
- - Eyelid Fold Consistency: New definition and techniques for representing diverse eyelid shapes in facial modeling
- - Scribble: Automatic generation of stylized 2D avatars from selfies
- - SimpleEgo: Solution for egocentric human pose estimation from downward-facing cameras on HMDs
Research Experience
- - Working in the Android XR team at Google.
- - Involved in the VoluMe project, which predicts 3D Gaussian reconstructions in real time from a single 2D webcam feed.
- - Part of the DAViD project, demonstrating the use of smaller but high-fidelity synthetic datasets for training models without loss in accuracy.
- - Contributor to the GASP project, proposing a method to overcome limitations of existing datasets.
- - Developer of 'Look Ma, no markers!', a technique for marker-free, high-quality reconstruction of the complete human body.
- - Participant in the Hairmony project, developing a method to predict a person's hairstyle from a single image.
- - Researcher on eyelid fold consistency, proposing a new definition and implementing geometric processing techniques.
- - Creator of Scribble, an application that automatically generates stylized 2D avatars from selfies.
- - Member of the SimpleEgo project, addressing egocentric human pose estimation from downward-facing cameras on HMDs.
Background
- Joining the Android XR org at Google as a research engineer. Will be working on synthetic human data for on-device sensing and supporting generative AI applications.
