Research Area
Year Published

631 Results

October 8, 2018

Sharding the Shards: Managing Datastore Locality at Scale with Akkio

USENIX Symposium on Operating Systems Design and Implementation (OSDI)

Akkio is a locality management service layered between client applications and distributed datastore systems. It determines how and when to migrate data to reduce response times and resource usage. Akkio primarily targets multi-datacenter geo-distributed datastore systems.

By: Muthukaruppan Annamalai, Kaushik Ravichandran, Harish Srinivas, Igor Zinkovsky, Luning Pan, Tony Savor, David Nagle, Michael Stumm
October 1, 2018

The effects of natural scene statistics on text readability in additive displays

Human Factors and Ergonomics Society

The minimum contrast needed for optimal text readability with additive displays (e.g. AR devices) will depend on the spatial structure of the background and text. Natural scenes and text follow similar spectral patterns. Therefore, natural scenes can mask low contrast text – making it difficult to read. In a set of experiments, we determine the minimum viable contrast for readability on an additive display.

By: Daryn R. Blanc-Goldhammer, Kevin J. MacKenzie
Areas: AR/VR
October 1, 2018

Low-Dose-Rate Cobalt-60 Testing Results for Kaman KD-5100 Differential Inductive Position Measuring Systems

Journal, IEEE Radiation Effects Data Workshop (REDW)

We report 60Co gamma radiation testing of a Kaman KD-5100 position measuring system to a total ionizing dose of 10 kRad(Si) at a rate of 5 mRad(Si)/s.

By: Bart McGuyer, Randall Milanowski, Slaven Moro
September 23, 2018

From Start-ups to Scale-ups: Opportunities and Open Problems for Static and Dynamic Program Analysis

IEEE International Working Conference on Source Code Analysis and Manipulation (SCAM)

This paper describes some of the challenges and opportunities when deploying static and dynamic analysis at scale, drawing on the authors’ experience with the Infer and Sapienz Technologies at Facebook, each of which started life as a research-led start-up that was subsequently deployed at scale, impacting billions of people worldwide.

By: Mark Harman, Peter O'Hearn
September 12, 2018

NAM: Non-Adversarial Unsupervised Domain Mapping

European Conference on Computer Vision (ECCV)

Several methods were recently proposed for the task of translating images between domains without prior knowledge in the form of correspondences. The existing methods apply adversarial learning to ensure that the distribution of the mapped source domain is indistinguishable from the target domain, which suffers from known stability issues. In addition, most methods rely heavily on “cycle” relationships between the domains, which enforce a one-to-one mapping. In this work, we introduce an alternative method: Non-Adversarial Mapping (NAM), which separates the task of target domain generative modeling from the cross-domain mapping task.

By: Yedid Hoshen, Lior Wolf
September 10, 2018

Value-aware Quantization for Training and Inference of Neural Networks

European Conference on Computer Vision (ECCV)

We propose a novel value-aware quantization which applies aggressively reduced precision to the majority of data while separately handling a small amount of large values in high precision, which reduces total quantization errors under very low precision.

By: Eunhyeok Park, Sungjoo Yoo, Peter Vajda
September 10, 2018

Predicting Future Instance Segmentation by Forecasting Convolutional Features

European Conference on Computer Vision (ECCV)

Anticipating future events is an important prerequisite towards intelligent behavior. Video forecasting has been studied as a proxy task towards this goal. Recent work has shown that to predict semantic segmentation of future frames, forecasting at the semantic level is more effective than forecasting RGB frames and then segmenting these. In this paper we consider the more challenging problem of future instance segmentation, which additionally segments out individual objects.

By: Pauline Luc, Camille Couprie, Yann LeCun, Jakob Verbeek
September 10, 2018

Dense Pose Transfer

European Conference on Computer Vision (ECCV)

In this work we integrate ideas from surface-based modeling with neural synthesis: we propose a combination of surface-based pose estimation and deep generative models that allows us to perform accurate pose transfer, i.e. synthesize a new image of a person based on a single image of that person and the image of a pose donor.

By: Natalia Neverova, Riza Alp Guler, Iasonas Kokkinos
September 9, 2018

Multi-Fiber Networks for Video Recognition

European Conference on Computer Vision (ECCV)

In this paper, we aim to reduce the computational cost of spatio-temporal deep neural networks, making them run as fast as their 2D counterparts while preserving state-of-the-art accuracy on video recognition benchmarks.

By: Yunpeng Chen, Yannis Kalantidis, Jianshu Li, Shuicheng Yan, Jiashi Feng