arXiv subjects

Yu Xiang

Publications and source records attributed to Yu Xiang.

At least 145 records · Page 8Linked to original sources

arXiv2018

Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objects

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 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. We show that for this problem the reality gap can be successfully spanned by a simple combination of domain randomized and photorealistic data. Using synthetic data generated in this manner, we introduce a one-shot deep neural network that is able to perform competitively against a state-of-the-art network trained on a combination of real and synthetic data. To our knowledge, this is the first deep network trained only on synthetic data that is able to achieve state-of-the-art performance on 6-DoF object pose estimation. Our network also generalizes better to novel environments including extreme lighting conditions, for which we show qualitative results. Using this network we demonstrate a real-time system estimating object poses with sufficient accuracy for real-world semantic grasping of known household objects in clutter by a real robot.

cs.RO↗

arXiv2018

FastTrack: Minimizing Stalls for CDN-based Over-the-top Video Streaming Systems

Traffic for internet video streaming has been rapidly increasing and is further expected to increase with the higher definition videos and IoT applications, such as 360 degree videos and augmented virtual reality applications. While efficient management of heterogeneous cloud resources to optimize the quality of experience is important, existing work in this problem space often left out important factors. In this paper, we present a model for describing a today's representative system architecture for video streaming applications, typically composed of a centralized origin server and several CDN sites. Our model comprehensively considers the following factors: limited caching spaces at the CDN sites, allocation of CDN for a video request, choice of different ports from the CDN, and the central storage and bandwidth allocation. With the model, we focus on minimizing a performance metric, stall duration tail probability (SDTP), and present a novel, yet efficient, algorithm to solve the formulated optimization problem. The theoretical bounds with respect to the SDTP metric are also analyzed and presented. Our extensive simulation results demonstrate that the proposed algorithms can significantly improve the SDTP metric, compared to the baseline strategies. Small-scale video streaming system implementation in a real cloud environment further validates our results.

cs.NI↗

arXiv2018

PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes

Estimating the 6D pose of known objects is important for robots to interact with the real world. The problem is challenging due to the variety of objects as well as the complexity of a scene caused by clutter and occlusions between objects. In this work, we introduce PoseCNN, a new Convolutional Neural Network for 6D object pose estimation. PoseCNN estimates the 3D translation of an object by localizing its center in the image and predicting its distance from the camera. The 3D rotation of the object is estimated by regressing to a quaternion representation. We also introduce a novel loss function that enables PoseCNN to handle symmetric objects. In addition, we contribute a large scale video dataset for 6D object pose estimation named the YCB-Video dataset. Our dataset provides accurate 6D poses of 21 objects from the YCB dataset observed in 92 videos with 133,827 frames. We conduct extensive experiments on our YCB-Video dataset and the OccludedLINEMOD dataset to show that PoseCNN is highly robust to occlusions, can handle symmetric objects, and provide accurate pose estimation using only color images as input. When using depth data to further refine the poses, our approach achieves state-of-the-art results on the challenging OccludedLINEMOD dataset. Our code and dataset are available at https://rse-lab.cs.washington.edu/projects/posecnn/.

cs.CV↗

arXiv2018

Recurrent Autoregressive Networks for Online Multi-Object Tracking

The main challenge of online multi-object tracking is to reliably associate object trajectories with detections in each video frame based on their tracking history. In this work, we propose the Recurrent Autoregressive Network (RAN), a temporal generative modeling framework to characterize the appearance and motion dynamics of multiple objects over time. The RAN couples an external memory and an internal memory. The external memory explicitly stores previous inputs of each trajectory in a time window, while the internal memory learns to summarize long-term tracking history and associate detections by processing the external memory. We conduct experiments on the MOT 2015 and 2016 datasets to demonstrate the robustness of our tracking method in highly crowded and occluded scenes. Our method achieves top-ranked results on the two benchmarks.

cs.CV↗

arXiv2018

Pseudo-Wigner Matrices

We consider the problem of generating pseudo-random matrices based on the similarity of their spectra to Wigner's semicircular law. We introduce the notion of an r-independent pseudo-Wigner matrix ensemble and prove closeness of the spectra of its matrices to the semicircular density in the Kolmogorov distance. We give an explicit construction of a family of N by N pseudo-Wigner ensembles using dual BCH codes and show that the Kolmogorov complexity of the obtained matrices is of the order of log(N) bits for a fixed designed Kolmogorov distance precision. We compare our construction to the quasi-random graphs introduced by Chung, Graham and Wilson and demonstrate that the pseudo-Wigner matrices pass stronger randomness tests than the adjacency matrices of these graphs (lifted by the mapping 0 -> 1 and 1 -> -1) do. Finally, we provide numerical simulations verifying our theoretical results.

cs.IT↗

arXiv2018

Symmetric Pseudo-Random Matrices

We consider the problem of generating symmetric pseudo-random sign (+/-1) matrices based on the similarity of their spectra to Wigner's semicircular law. Using binary m-sequences (Golomb sequences) of lengths n=2^m-1, we give a simple explicit construction of circulant n by n sign matrices and show that their spectra converge to the semicircular law when n grows. The Kolmogorov complexity of the proposed matrices equals to that of Golomb sequences and is at most 2log(n) bits.

math.PR↗

arXiv2018

Multipartite Einstein-Podolsky-Rosen steering sharing with separable states

Distribution of quantum correlations among remote users is a key procedure underlying many quantum information technologies. Einstein-Podolsky-Rosen steering, which is one kind of such correlations stronger than entanglement, has been identified as a resource for secure quantum networks. We show that this resource can be established between two and even more distant parties by transmission of a system being separable from all the parties. For the case with two parties, we design a protocol allowing to distribute one-way Gaussian steering between them via a separable carrier; the obtained steering can be used subsequently for one-sided device-independent (1sDI) quantum key distribution. Further, we extend the protocol to three parties, a scenario which exhibits richer steerability properties including one-to-multimode steering and collective steering, and which can be used for 1sDI quantum secret sharing. All the proposed steering distribution protocols can be implemented with squeezed states, beam splitters and displacements, and thus they can be readily realized experimentally. Our findings reveal that not only entanglement but even steering can be distributed via communication of a separable system. Viewed from a different perspective, the present protocols also demonstrate that one can switch multipartite states between different steerability classes by operations on parts of the states.

quant-ph↗

arXiv2017

Investigating Einstein-Podolsky-Rosen steering of continuous variable bipartite states by non-Gaussian pseudospin measurements

EPR steering is an asymmetric form of correlations which is intermediate between quantum entanglement and Bell nonlocality, and can be exploited for quantum communication with one untrusted party. In particular, steering of continuous variable Gaussian states has been extensively studied as a manifestation of the EPR paradox. While most of these studies focused on quadrature measurements for steering detection, two recent works revealed that there exist Gaussian states which are only steerable by non-Gaussian measurements. In this paper we perform a systematic investigation of EPR steering of bipartite Gaussian states by pseudospin measurements, complementing and extending previous findings. We first derive the density matrix elements of two-mode squeezed thermal states in the Fock basis, which may be of independent interest. We then use such a representation to investigate steering of these states as detected by a nonlinear criterion, based on second moments of the pseudospin correlation matrix. This analysis reveals previously unexplored regimes where non-Gaussian measurements are more effective than Gaussian ones to witness steering of Gaussian states in the presence of local noise. We further consider an alternative set of pseudospin observables, whose expectation value can be expressed compactly in terms of Wigner functions for all two-mode Gaussian states. However, according to the adopted criterion, these observables are found to be always less sensitive than Gaussian observables for steering detection. Finally, we investigate continuous variable Werner states, which are non-Gaussian mixtures of Gaussian states, and find that pseudospin measurements are always more effective than Gaussian ones to reveal their steerability. Our results provide useful insights on the role of non-Gaussian measurements in characterizing quantum correlations of Gaussian and non-Gaussian states.

quant-ph↗

arXiv2017

Sprout: A functional caching approach to minimize service latency in erasure-coded storage

Modern distributed storage systems often use erasure codes to protect against disk and node failures to increase reliability, while trying to meet the latency requirements of the applications and clients. Storage systems may have caches at the proxy or client ends in order to reduce the latency. In this paper, we consider a novel caching framework with erasure code called functional caching. Functional Caching involves using erasure-coded chunks in the cache such that the code formed by the chunks in storage nodes and cache combined are maximal-distance-separable (MDS) erasure codes. Based on the arrival rates of different files, placement of file chunks on the servers, and service time distribution of storage servers, an optimal functional caching placement and the access probabilities of the file request from different disks are considered. The proposed algorithm gives significant latency improvement in both simulations and a prototyped solution in an open-source, cloud storage deployment.

cs.DC↗

arXiv2017

Demonstration of monogamy relations for Einstein-Podolsky-Rosen steering in Gaussian cluster states

Understanding how quantum resources can be quantified and distributed over many parties has profound applications in quantum communication. As one of the most intriguing features of quantum mechanics, Einstein-Podolsky-Rosen (EPR) steering is a useful resource for secure quantum networks. By reconstructing the covariance matrix of a continuous variable four-mode square Gaussian cluster state subject to asymmetric loss, we quantify the amount of bipartite steering with a variable number of modes per party, and verify recently introduced monogamy relations for Gaussian steerability, which establish quantitative constraints on the security of information shared among different parties. We observe a very rich structure for the steering distribution, and demonstrate one-way EPR steering of the cluster state under Gaussian measurements, as well as one-to-multi-mode steering. Our experiment paves the way for exploiting EPR steering in Gaussian cluster states as a valuable resource for multiparty quantum information tasks.

quant-ph↗

arXiv2017

DA-RNN: Semantic Mapping with Data Associated Recurrent Neural Networks

3D scene understanding is important for robots to interact with the 3D world in a meaningful way. Most previous works on 3D scene understanding focus on recognizing geometrical or semantic properties of the scene independently. In this work, we introduce Data Associated Recurrent Neural Networks (DA-RNNs), a novel framework for joint 3D scene mapping and semantic labeling. DA-RNNs use a new recurrent neural network architecture for semantic labeling on RGB-D videos. The output of the network is integrated with mapping techniques such as KinectFusion in order to inject semantic information into the reconstructed 3D scene. Experiments conducted on a real world dataset and a synthetic dataset with RGB-D videos demonstrate the ability of our method in semantic 3D scene mapping.

cs.CV↗

arXiv2017

Subcategory-aware Convolutional Neural Networks for Object Proposals and Detection

In CNN-based object detection methods, region proposal becomes a bottleneck when objects exhibit significant scale variation, occlusion or truncation. In addition, these methods mainly focus on 2D object detection and cannot estimate detailed properties of objects. In this paper, we propose subcategory-aware CNNs for object detection. We introduce a novel region proposal network that uses subcategory information to guide the proposal generating process, and a new detection network for joint detection and subcategory classification. By using subcategories related to object pose, we achieve state-of-the-art performance on both detection and pose estimation on commonly used benchmarks.

cs.CV↗

arXiv2017

Multipartite Gaussian steering: monogamy constraints and quantum cryptography applications

We derive laws for the distribution of quantum steering among different parties in multipartite Gaussian states under Gaussian measurements. We prove that a monogamy relation akin to the generalized Coffman-Kundu-Wootters inequality holds quantitatively for a recently introduced measure of Gaussian steering. We then define the residual Gaussian steering, stemming from the monogamy inequality, as an indicator of collective steering-type correlations. For pure three-mode Gaussian states, the residual acts a quantifier of genuine multipartite steering, and is interpreted operationally in terms of the guaranteed key rate in the task of secure quantum secret sharing. Optimal resource states for the latter protocol are identified, and their possible experimental implementation discussed. Our results pin down the role of multipartite steering for quantum communication.

quant-ph↗

arXiv2017

Unconditional security of entanglement-based continuous-variable quantum secret sharing

The need for secrecy and security is essential in communication. Secret sharing is a conventional protocol to distribute a secret message to a group of parties, who cannot access it individually but need to cooperate in order to decode it. While several variants of this protocol have been investigated, including realizations using quantum systems, the security of quantum secret sharing schemes still remains unproven almost two decades after their original conception. Here we establish an unconditional security proof for continuous variable entanglement-based quantum secret sharing schemes, in the limit of asymptotic keys and for an arbitrary number of players. We tackle the problem by resorting to the recently developed one-sided device-independent approach to quantum key distribution. We demonstrate theoretically the feasibility of our scheme, which can be implemented by Gaussian states and homodyne measurements, with no need for ideal single-photon sources or quantum memories. Our results contribute to validating quantum secret sharing as a viable primitive for quantum technologies.

quant-ph↗

arXiv2016

Multiple Change Point Analysis: Fast Implementation And Strong Consistency

One of the main challenges in identifying structural changes in stochastic processes is to carry out analysis for time series with dependency structure in a computationally tractable way. Another challenge is that the number of true change points is usually unknown, requiring a suitable model selection criterion to arrive at informative conclusions. To address the first challenge, we model the data generating process as a segment-wise autoregression, which is composed of several segments (time epochs), each of which modeled by an autoregressive model. We propose a multi-window method that is both effective and efficient for discovering the structural changes. The proposed approach was motivated by transforming a segment-wise autoregression into a multivariate time series that is asymptotically segment-wise independent and identically distributed. To address the second challenge, we derive theoretical guarantees for (almost surely) selecting the true number of change points of segment-wise independent multivariate time series. Specifically, under mild assumptions, we show that a Bayesian Information Criterion (BIC)-like criterion gives a strongly consistent selection of the optimal number of change points, while an Akaike Information Criterion (AIC)-like criterion cannot. Finally, we demonstrate the theory and strength of the proposed algorithms by experiments on both synthetic and real-world data, including the Eastern US temperature data and the El Nino data from 1854 to 2015. The experiment leads to some interesting discoveries about temporal variability of the summer-time temperature over the Eastern US, and about the most dominant factor of ocean influence on climate.

stat.ME↗

arXiv2016

Differentiated latency in data center networks with erasure coded files through traffic engineering

This paper proposes an algorithm to minimize weighted service latency for different classes of tenants (or service classes) in a data center network where erasure-coded files are stored on distributed disks/racks and access requests are scattered across the network. Due to limited bandwidth available at both top-of-the-rack and aggregation switches and tenants in different service classes need differentiated services, network bandwidth must be apportioned among different intra- and inter-rack data flows for different service classes in line with their traffic statistics. We formulate this problem as weighted queuing and employ a class of probabilistic request scheduling policies to derive a closed-form upper-bound of service latency for erasure-coded storage with arbitrary file access patterns and service time distributions. The result enables us to propose a joint weighted latency (over different service classes) optimization over three entangled "control knobs": the bandwidth allocation at top-of-the-rack and aggregation switches for different service classes, dynamic scheduling of file requests, and the placement of encoded file chunks (i.e., data locality). The joint optimization is shown to be a mixed-integer problem. We develop an iterative algorithm which decouples and solves the joint optimization as 3 sub-problems, which are either convex or solvable via bipartite matching in polynomial time. The proposed algorithm is prototyped in an open-source, distributed file system, {\em Tahoe}, and evaluated on a cloud testbed with 16 separate physical hosts in an Openstack cluster using 48-port Cisco Catalyst switches. Experiments validate our theoretical latency analysis and show significant latency reduction for diverse file access patterns. The results provide valuable insights on designing low-latency data center networks with erasure coded storage.

cs.DC↗

arXiv2015

Deep Metric Learning via Lifted Structured Feature Embedding

Learning the distance metric between pairs of examples is of great importance for learning and visual recognition. With the remarkable success from the state of the art convolutional neural networks, recent works have shown promising results on discriminatively training the networks to learn semantic feature embeddings where similar examples are mapped close to each other and dissimilar examples are mapped farther apart. In this paper, we describe an algorithm for taking full advantage of the training batches in the neural network training by lifting the vector of pairwise distances within the batch to the matrix of pairwise distances. This step enables the algorithm to learn the state of the art feature embedding by optimizing a novel structured prediction objective on the lifted problem. Additionally, we collected Online Products dataset: 120k images of 23k classes of online products for metric learning. Our experiments on the CUB-200-2011, CARS196, and Online Products datasets demonstrate significant improvement over existing deep feature embedding methods on all experimented embedding sizes with the GoogLeNet network.

cs.CV↗

arXiv2015

A Coarse-to-Fine Model for 3D Pose Estimation and Sub-category Recognition

Despite the fact that object detection, 3D pose estimation, and sub-category recognition are highly correlated tasks, they are usually addressed independently from each other because of the huge space of parameters. To jointly model all of these tasks, we propose a coarse-to-fine hierarchical representation, where each level of the hierarchy represents objects at a different level of granularity. The hierarchical representation prevents performance loss, which is often caused by the increase in the number of parameters (as we consider more tasks to model), and the joint modelling enables resolving ambiguities that exist in independent modelling of these tasks. We augment PASCAL3D+ dataset with annotations for these tasks and show that our hierarchical model is effective in joint modelling of object detection, 3D pose estimation, and sub-category recognition.

cs.CV↗