Georgia Gkioxari
Georgia Gkioxari completed a Berkeley doctorate in 2016 after studying at NTUA. She worked at Meta’s FAIR team before joining Caltech, where she became an assistant professor in 2023. She co-authored Mask R-CNN in 2017, extending object detection with a branch that predicts instance masks. Her work includes PyTorch3D and Omni3D. The group’s current research studies 2D and 3D perception and spatial reasoning.
- Works
- 76
- Citations
- 39,952
- h-index
- 33
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2010–2016
Training in visual research
She earned a NTUA diplom in 2010 and a Berkeley doctorate in 2016. Jitendra Malik was her doctoral adviser, and she later conducted research at FAIR.
-
2017
Masks for individual objects
Gkioxari co-authored Mask R-CNN with Kaiming He, Piotr Dollár, and Ross Girshick. First submitted in March 2017, the method adds a mask-prediction branch to Faster R-CNN. It supports object instance segmentation by detecting an object and predicting the pixels that belong to it.
-
2023
A Caltech appointment
She was a visiting associate in 2022 and became an assistant professor in 2023. She worked at FAIR before joining Caltech.
-
As of October 2026
3D tools and perception research
She contributed to PyTorch3D, an open-source library for 3D deep learning, and Omni3D, a 2023 benchmark and model for 3D object detection. Her group studies visual perception and spatial reasoning.
Sources and credits
Photo: www.eas.caltech.edu; creator not stated. License: Not stated by source. Source: eas.caltech.edu. Cropped to a square and resized. Ask us to remove or replace this photo.