Terrance DeVries
YOU?
Author Swipe
View article: Unconstrained Scene Generation with Locally Conditioned Radiance Fields
Unconstrained Scene Generation with Locally Conditioned Radiance Fields Open
We tackle the challenge of learning a distribution over complex, realistic, indoor scenes. In this paper, we introduce Generative Scene Networks (GSN), which learns to decompose scenes into a collection of many local radiance fields that c…
View article: The GIST and RIST of Iterative Self-Training for Semi-Supervised Segmentation
The GIST and RIST of Iterative Self-Training for Semi-Supervised Segmentation Open
We consider the task of semi-supervised semantic segmentation, where we aim to produce pixel-wise semantic object masks given only a small number of human-labeled training examples. We focus on iterative self-training methods in which we e…
View article: Building LEGO Using Deep Generative Models of Graphs
Building LEGO Using Deep Generative Models of Graphs Open
Generative models are now used to create a variety of high-quality digital artifacts. Yet their use in designing physical objects has received far less attention. In this paper, we advocate for the construction toy, LEGO, as a platform for…
View article: Instance Selection for GANs
Instance Selection for GANs Open
Recent advances in Generative Adversarial Networks (GANs) have led to their widespread adoption for the purposes of generating high quality synthetic imagery. While capable of generating photo-realistic images, these models often produce u…
View article: ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component\n Analysis
ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component\n Analysis Open
We consider the problem of distance metric learning (DML), where the task is\nto learn an effective similarity measure between images. We revisit ProxyNCA\nand incorporate several enhancements. We find that low temperature scaling is a\npe…
View article: On the Evaluation of Conditional GANs
On the Evaluation of Conditional GANs Open
Conditional Generative Adversarial Networks (cGANs) are finding increasingly widespread use in many application domains. Despite outstanding progress, quantitative evaluation of such models often involves multiple distinct metrics to asses…
View article: Does Object Recognition Work for Everyone?
Does Object Recognition Work for Everyone? Open
The paper analyzes the accuracy of publicly available object-recognition systems on a geographically diverse dataset. This dataset contains household items and was designed to have a more representative geographical coverage than commonly …
View article: Leveraging Uncertainty Estimates for Predicting Segmentation Quality
Leveraging Uncertainty Estimates for Predicting Segmentation Quality Open
The use of deep learning for medical imaging has seen tremendous growth in the research community. One reason for the slow uptake of these systems in the clinical setting is that they are complex, opaque and tend to fail silently. Outside …
View article: Learning Confidence for Out-of-Distribution Detection in Neural Networks
Learning Confidence for Out-of-Distribution Detection in Neural Networks Open
Modern neural networks are very powerful predictive models, but they are often incapable of recognizing when their predictions may be wrong. Closely related to this is the task of out-of-distribution detection, where a network must determi…
View article: Improved Regularization of Convolutional Neural Networks with Cutout
Improved Regularization of Convolutional Neural Networks with Cutout Open
Convolutional neural networks are capable of learning powerful representational spaces, which are necessary for tackling complex learning tasks. However, due to the model capacity required to capture such representations, they are often su…
View article: LesionSeg: Semantic segmentation of skin lesions using Deep Convolutional Neural Network
LesionSeg: Semantic segmentation of skin lesions using Deep Convolutional Neural Network Open
We present a method for skin lesion segmentation for the ISIC 2017 Skin Lesion Segmentation Challenge. Our approach is based on a Fully Convolutional Network architecture which is trained end to end, from scratch, on a limited dataset. Our…
View article: Skin Lesion Classification Using Deep Multi-scale Convolutional Neural Networks
Skin Lesion Classification Using Deep Multi-scale Convolutional Neural Networks Open
We present a deep learning approach to the ISIC 2017 Skin Lesion Classification Challenge using a multi-scale convolutional neural network. Our approach utilizes an Inception-v3 network pre-trained on the ImageNet dataset, which is fine-tu…
View article: Dataset Augmentation in Feature Space
Dataset Augmentation in Feature Space Open
Dataset augmentation, the practice of applying a wide array of domain-specific transformations to synthetically expand a training set, is a standard tool in supervised learning. While effective in tasks such as visual recognition, the set …