Using synthetic data for training deep neural networks for robotic manipulation holds the promise of an almost unlimited amount of pre-labeled training data, generated safely out of harm's way. One of the key challenges of synthetic data, to date, has been to bridge the so-called \emph{reality gap}, so that networks trained on synthetic data operate correctly when exposed to real-world data. We explore the reality gap in the context of 6-DoF pose estimation of known objects from a single RGB image.

Superpixels provide an efficient low/mid-level representation of image data, which greatly reduces the number of image primitives for subsequent vision tasks. While various superpixel computation models exist, they are not differentiable, making them difficult to integrate into otherwise end-to-end trainable deep neural networks. In this work, we develop a new differentiable model for superpixel sampling that better leverages deep networks for learning superpixel segmentation.

Application execution on safety-critical and high-performance computer systems must be resilient to transient errors. As GPUs become more pervasive in such systems, they must supplement ECC/parity for major storage structures with reliability techniques that cover more of the GPU hardware logic. Instruction duplication has been explored for CPU resilience; however, it has never been studied in the context of GPUs, and it is unclear whether the performance and design choices it presents make it a feasible GPU solution.

The design challenges of see-through near-eye displays can be mitigated by specializing an augmented reality device for a particular application. We present a novel optical design for augmented reality near-eye displays exploiting 3D stereolithography printing techniques to achieve similar characteristics to progressive prescription binoculars. We propose to manufacture inter-changeable optical components using 3D printing, leading to arbitrary shaped static projection screen surfaces that are adaptive to the targeted applications.

Geometric aliasing is a persistent challenge for real-time rendering. Hardware multisampling remains limited to 8 × , analytic coverage fails to capture correlated visibility samples, and spatial and temporal postfiltering primarily target edges of superpixel primitives. We describe a novel semi-analytic representation of coverage designed to make progress on geometric antialiasing for subpixel primitives and pixels containing many edges while handling correlated subpixel coverage.

We present an iterative overlap estimation technique to augment existing point cloud registration algorithms that can achieve high performance in difficult real-world situations where large pose displacement and non-overlapping geometry would otherwise cause traditional methods to fail. Our approach estimates overlapping regions through an iterative Expectation Maximization procedure that encodes the sensor field-of-view into the registration process.

We present two techniques to improve landmark localization in images from partially annotated datasets. Our primary goal is to leverage the common situation where precise landmark locations are only provided for a small data subset, but where class labels for classification or regression tasks related to the landmarks are more abundantly available. First, we propose the framework of sequential multitasking and explore it here through an architecture for landmark localization where training with class labels acts as an auxiliary signal to guide the landmark localization on unlabeled data.

Superpixel segmentation has been widely used in many computer vision tasks. Existing superpixel algorithms are mainly based on hand-crafted features, which often fail to preserve weak object boundaries. In this work, we leverage deep neural networks to facilitate extracting superpixels from images. We show a simple integration of deep features with existing superpixel algorithms does not result in better performance as these features do not model segmentation. Instead, we propose a segmentation-aware affinity learning approach for superpixel segmentation.

In this paper, we strive to answer two questions: What is the current state of 3D hand pose estimation from depth images? And, what are the next challenges that need to be tackled? Following the successful Hands In the Million Challenge (HIM2017), we investigate the top 10 state-of-the-art methods on three tasks: single frame 3D pose estimation, 3D hand tracking, and hand pose estimation during object interaction. We analyze the performance of different CNN structures with regard to hand shape, joint visibility, view point and articulation distributions.