Terrance DeVries

I am a research scientist at Luma AI, where I work on machine learning and computer vision for 3D capture.

I received my PhD from the University of Guelph, where I was advised by Graham Taylor. I interned with Josh Susskind's research group at Apple from summer 2020 to fall 2021. In winter 2019 I worked with Michal Drodzal as an intern at Facebook AI Research, and in fall 2018 I interned with Laurens van der Maaten, also at Facebook AI Research. I did my undergad in mechanical engineering, at the University of Guelph.

Email  /  GitHub  /  Google Scholar  /  LinkedIn

profile photo

Research

I'm interested in computer vision and machine learning, especially image generation and 3D scene synthesis. My goal is to create models that can generate photorealistic and explorable 3D worlds. Representative papers are highlighted.

project image

Unconstrained Scene Generation with Locally Conditioned Radiance Fields


Terrance DeVries, Miguel Angel Bautista, Nitish Srivastava, Graham W. Taylor, Joshua M. Susskind
ICCV, 2021
paper / project page / code / blog

Generative model for learning radiance fields of realistic scenes. Allows for generating 3D scenes and exploring them with a freely moving camera.

project image

The GIST and RIST of Iterative Self-Training for Semi-Supervised Segmentation


Eu Wern Teh, Terrance DeVries, Brendan Duke, Ruowei Jiang, Parham Aarabi, Graham W. Taylor
arXiv, 2021
paper

Alternating between training on human labeled data and psuedo-labeled data stabilizes iterative self-training, enabling significant improvement in semi-supervised segmentation quality.

project image

Building LEGO Using Deep Generative Models of Graphs


Rylee Thompson, Elahe Ghalebi, Terrance DeVries, Graham W. Taylor
NeurIPS ML4Eng Workshop, 2020
paper / code

Generative model for sequential assembly, demonstrated by building LEGO structures piece-by-piece.

project image

Instance Selection for GANs


Terrance DeVries, Michal Drozdzal, Graham W. Taylor
NeurIPS, 2020
paper / code

Removing outliers from the training set trades GAN sample diversity for higher quality samples, faster training, and reduced model capacity requirements.

project image

ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis


Eu Wern Teh, Terrance DeVries, Graham W. Taylor
ECCV, 2020
paper / code

ProxyNCA surpasses state-of-the-art metric learning methods when updated with a bag of tricks.

project image

On the Evaluation of Conditional GANs


Terrance DeVries, Adriana Romero, Luis Pineda, Graham W. Taylor, Michal Drozdzal
arXiv, 2019
paper / code

Conditional generative models can be evaluated by measuring the distance between real and generated joint distributions of images and conditioning.

project image

Does Object Recognition Work for Everyone?


Terrance DeVries*, Ishan Misra*, Changhan Wang*, Laurens van der Maaten
CVPR Workshop on Computer Vision for Global Challenges (CV4GC), 2019
paper / blog

Publicly available object-recognition systems perform relatively poorly on household items in countries with a low household income.

project image

Leveraging Uncertainty Estimates for Predicting Segmentation Quality


Terrance DeVries, Graham W. Taylor
arXiv, 2018
paper

Uncertainty estimates from segmentation networks can be used to predict the quality of the segmentations.

project image

Learning Confidence for Out-of-Distribution Detection in Neural Networks


Terrance DeVries, Graham W. Taylor
arXiv, 2018
paper / code

Training a confidence estimation branch on classification networks enables identification of out-of-distribution examples.

project image

Improved Regularization of Convolutional Neural Networks with Cutout


Terrance DeVries, Graham W. Taylor
arXiv, 2017
paper / code

Randomly masking square regions of pixels from training images is an effective form of data augmentation.

project image

Dataset Augmentation in Feature Space


Terrance DeVries, Graham W. Taylor
ICLR Workshop, 2017
paper

Datasets can be augmented by interpolating or extrapolating between datapoints in feature space to improve classification accuracy.

project image

Multi-task Learning of Facial Landmarks and Expression


Terrance DeVries*, Kumar Biswaranjan*, Graham W. Taylor
CRV, 2014
paper

Facial expression recognition accuracy can be improved if the model is also trained to identify facial landmarks.





Design and source code from Leonid Keselman's website, which is based on Jon Barron's website