Noise2Noise: Learning Image Restoration without Clean Data

We apply basic statistical reasoning to signal reconstruction by machine learning — learning to map corrupted observations to clean signals — with a simple and powerful conclusion: under certain common circumstances, it is possible to learn to restore signals without ever observing clean ones, at performance close or equal to training using clean exemplars. We show applications in photographic noise removal, denoising of synthetic Monte Carlo images, and reconstruction of MRI scans from undersampled inputs, all based on only observing corrupted data.

Light-weight Head Pose Invariant Gaze Tracking

Unconstrained remote gaze tracking using off-the-shelf cameras is a challenging problem. Recently, promising algorithms for appearance-based gaze estimation using convolutional neural networks (CNN) have been proposed. Improving their robustness to various confounding factors including variable head pose, subject identity, illumination and image quality remain open problems. In this work, we study the effect of variable head pose on machine learning regressors trained to estimate gaze direction.

A Modular Digital VLSI Flow for High-Productivity SoC Design

A high-productivity digital VLSI flow for designing complex SoCs is presented. The flow includes high-level synthesis tools, an object-oriented library of synthesizable SystemC and C++ components, and a modular VLSI physical design approach based on fine-grained globally asynchronous locally synchronous (GALS) clocking. The flow was demonstrated on a 16nm FinFET testchip targeting machine learning and computer vision.

Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization

We present a system for training deep neural networks for object detection using synthetic images. To handle the variability in real-world data, the system relies upon the technique of domain randomization, in which the parameters of the simulator—such as lighting, pose, object textures, etc.—are randomized in non-realistic ways to force the neural network to learn the essential features of the object of interest.

On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach

We revisit the problem of visual depth estimation in the context of autonomous vehicles. Despite the progress on monocular depth estimation in recent years, we show that the gap between monocular and stereo depth accuracy remains large---a particularly relevant result due to the prevalent reliance upon monocular cameras by vehicles that are expected to be self-driving. We argue that the challenges of removing this gap are significant, owing to fundamental limitations of monocular vision. As a result, we focus our efforts on depth estimation by stereo.

Falling Things: A Synthetic Dataset for 3D Object Detection and Pose Estimation

We present a new dataset, called Falling Things (FAT), for advancing the state-of-the-art in object detection and 3D pose estimation in the context of robotics.

SPLATNet: Sparse Lattice Networks for Point Cloud Processing

We present a network architecture for processing point clouds that directly operates on a collection of points represented as a sparse set of samples in a high-dimensional lattice. Naively applying convolutions on this lattice scales poorly, both in terms of memory and computational cost, as the size of the lattice increases. Instead, our network uses sparse bilateral convolutional layers as building blocks.

Combining Analytic Direct Illumination and Stochastic Shadows

In this paper, we propose a ratio estimator of the direct-illumination equation that allows us to combine analytic illumination techniques with stochastic raytraced shadows while maintaining correctness. Our main contribution is to show that the shadowed illumination can be split into the product of the unshadowed illumination and the illumination-weighted shadow.

Modeling Soft Error Propagation in Programs

As technology scales to lower feature sizes, devices become more susceptible to soft errors. Soft errors can lead to silent data corruptions (SDCs), seriously compromising the reliability of a system. Traditional hardware-only techniques to avoid SDCs are energy hungry, and hence not suitable for commodity systems. Researchers have proposed selective software-based protection techniques to tolerate hardware faults at lower costs.

Riemannian Motion Policies

A new mathematical framework called Riemannian Motion Policies (RMPs) shapes a robot’s behavior. We derive optimal and practical tools for intuitively constructing policies, demonstrate the framework’s flexibility for distributed computation, use it to unify many previously distinct motion generation techniques, and demonstrate its performance on three dual arm manipulation platforms in both simulation and reality.