arXiv subjects

Song Liu

Publications and source records attributed to Song Liu.

214 records · Page 12Linked to original sources

arXiv2016

Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks

This paper addresses the problem of continuous gesture recognition from sequences of depth maps using convolutional neutral networks (ConvNets). The proposed method first segments individual gestures from a depth sequence based on quantity of movement (QOM). For each segmented gesture, an Improved Depth Motion Map (IDMM), which converts the depth sequence into one image, is constructed and fed to a ConvNet for recognition. The IDMM effectively encodes both spatial and temporal information and allows the fine-tuning with existing ConvNet models for classification without introducing millions of parameters to learn. The proposed method is evaluated on the Large-scale Continuous Gesture Recognition of the ChaLearn Looking at People (LAP) challenge 2016. It achieved the performance of 0.2655 (Mean Jaccard Index) and ranked $3^{rd}$ place in this challenge.

cs.CV↗

arXiv2016

Structure Learning of Partitioned Markov Networks

We learn the structure of a Markov Network between two groups of random variables from joint observations. Since modelling and learning the full MN structure may be hard, learning the links between two groups directly may be a preferable option. We introduce a novel concept called the \emph{partitioned ratio} whose factorization directly associates with the Markovian properties of random variables across two groups. A simple one-shot convex optimization procedure is proposed for learning the \emph{sparse} factorizations of the partitioned ratio and it is theoretically guaranteed to recover the correct inter-group structure under mild conditions. The performance of the proposed method is experimentally compared with the state of the art MN structure learning methods using ROC curves. Real applications on analyzing bipartisanship in US congress and pairwise DNA/time-series alignments are also reported.

stat.ML↗

arXiv2016

Support Consistency of Direct Sparse-Change Learning in Markov Networks

We study the problem of learning sparse structure changes between two Markov networks $P$ and $Q$. Rather than fitting two Markov networks separately to two sets of data and figuring out their differences, a recent work proposed to learn changes \emph{directly} via estimating the ratio between two Markov network models. In this paper, we give sufficient conditions for \emph{successful change detection} with respect to the sample size $n_p, n_q$, the dimension of data $m$, and the number of changed edges $d$. When using an unbounded density ratio model we prove that the true sparse changes can be consistently identified for $n_p = Ω(d^2 \log \frac{m^2+m}{2})$ and $n_q = Ω({n_p^2})$, with an exponentially decaying upper-bound on learning error. Such sample complexity can be improved to $\min(n_p, n_q) = Ω(d^2 \log \frac{m^2+m}{2})$ when the boundedness of the density ratio model is assumed. Our theoretical guarantee can be applied to a wide range of discrete/continuous Markov networks.

stat.ML↗

arXiv2016

Creating Simplified 3D Models with High Quality Textures

This paper presents an extension to the KinectFusion algorithm which allows creating simplified 3D models with high quality RGB textures. This is achieved through (i) creating model textures using images from an HD RGB camera that is calibrated with Kinect depth camera, (ii) using a modified scheme to update model textures in an asymmetrical colour volume that contains a higher number of voxels than that of the geometry volume, (iii) simplifying dense polygon mesh model using quadric-based mesh decimation algorithm, and (iv) creating and mapping 2D textures to every polygon in the output 3D model. The proposed method is implemented in real-time by means of GPU parallel processing. Visualization via ray casting of both geometry and colour volumes provides users with a real-time feedback of the currently scanned 3D model. Experimental results show that the proposed method is capable of keeping the model texture quality even for a heavily decimated model and that, when reconstructing small objects, photorealistic RGB textures can still be reconstructed.

cs.GR↗

arXiv2016

Planogram Compliance Checking Based on Detection of Recurring Patterns

In this paper, a novel method for automatic planogram compliance checking in retail chains is proposed without requiring product template images for training. Product layout is extracted from an input image by means of unsupervised recurring pattern detection and matched via graph matching with the expected product layout specified by a planogram to measure the level of compliance. A divide and conquer strategy is employed to improve the speed. Specifically, the input image is divided into several regions based on the planogram. Recurring patterns are detected in each region respectively and then merged together to estimate the product layout. Experimental results on real data have verified the efficacy of the proposed method. Compared with a template-based method, higher accuracies are achieved by the proposed method over a wide range of products.

cs.CV↗

arXiv2015

Estimating Posterior Ratio for Classification: Transfer Learning from Probabilistic Perspective

Transfer learning assumes classifiers of similar tasks share certain parameter structures. Unfortunately, modern classifiers uses sophisticated feature representations with huge parameter spaces which lead to costly transfer. Under the impression that changes from one classifier to another should be ``simple'', an efficient transfer learning criteria that only learns the ``differences'' is proposed in this paper. We train a \emph{posterior ratio} which turns out to minimizes the upper-bound of the target learning risk. The model of posterior ratio does not have to share the same parameter space with the source classifier at all so it can be easily modelled and efficiently trained. The resulting classifier therefore is obtained by simply multiplying the existing probabilistic-classifier with the learned posterior ratio.

stat.ML↗

arXiv2015

Inverse Doppler Effects in Flute

Here we report the observation of the inverse Doppler effects in a flute. It is experimentally verified that, when there is a relative movement between the source and the observer, the inverse Doppler effect could be detected for all seven pitches of a musical scale produced by a flute. Higher tone is associated with a greater shift in frequency. The effect of the inverse frequency shift may provide new insights into why the flute, with its euphonious tone, has been popular for thousands of years in Asia and Europe.

physics.class-ph↗

arXiv2015

Modeling Relation Paths for Representation Learning of Knowledge Bases

Representation learning of knowledge bases (KBs) aims to embed both entities and relations into a low-dimensional space. Most existing methods only consider direct relations in representation learning. We argue that multiple-step relation paths also contain rich inference patterns between entities, and propose a path-based representation learning model. This model considers relation paths as translations between entities for representation learning, and addresses two key challenges: (1) Since not all relation paths are reliable, we design a path-constraint resource allocation algorithm to measure the reliability of relation paths. (2) We represent relation paths via semantic composition of relation embeddings. Experimental results on real-world datasets show that, as compared with baselines, our model achieves significant and consistent improvements on knowledge base completion and relation extraction from text.

cs.CL↗

arXiv2015

Experimental Observation of Reversed Doppler Effects in Acoustic Metamaterials

This paper reports an experimental observation of broadband reversed Doppler effects using an acoustic metamaterial with seven flute-like double-meta-molecule clusters. Simulations and experiments verify that this locally resonant acoustic metamaterial with simultaneous negative elastic modulus and mass density can realize negative refraction in a broad frequency range. The constructed metamaterial exhibits broadband reversed Doppler effects. The frequency shift increases continuously as the frequency increases. The assembling of double-meta-molecule clusters introduces a new direction in designing double-negative acoustic metamaterials in an arbitrarily broad frequency range and other various applications.

cond-mat.mes-hall↗

arXiv2015

Anomalous refraction based on acoustic metasurfaces with membranes

The investigation of metasurface, which is of great current interest, has opened up new degrees of freedom to research metamaterials. In this paper, we propose an ultrathin acoustic metasurface consisting of a series of structurally simple drum-like subunit cells. Eight units with different sizes aer selected to realize discrete transmisson phase shifts ranging from 0 to 360 degrees ar 3.5 KHz. The designed metasurface is capable of manipulating sound waves at will, including anomalous refraction and conversion of propagating waves into surface waves, which can be predicted precisely by the generalized Snell's law. It id worth noting that the total thickness of such planar metasurface is approximately 1/5 of working wavelength, which may be beneficial to the miniaturization and integration of acoustic equipment.

physics.class-ph↗

arXiv2014

Direct Learning of Sparse Changes in Markov Networks by Density Ratio Estimation

We propose a new method for detecting changes in Markov network structure between two sets of samples. Instead of naively fitting two Markov network models separately to the two data sets and figuring out their difference, we \emph{directly} learn the network structure change by estimating the ratio of Markov network models. This density-ratio formulation naturally allows us to introduce sparsity in the network structure change, which highly contributes to enhancing interpretability. Furthermore, computation of the normalization term, which is a critical bottleneck of the naive approach, can be remarkably mitigated. We also give the dual formulation of the optimization problem, which further reduces the computation cost for large-scale Markov networks. Through experiments, we demonstrate the usefulness of our method.

stat.ML↗

arXiv2013

Change-Point Detection in Time-Series Data by Relative Density-Ratio Estimation

The objective of change-point detection is to discover abrupt property changes lying behind time-series data. In this paper, we present a novel statistical change-point detection algorithm based on non-parametric divergence estimation between time-series samples from two retrospective segments. Our method uses the relative Pearson divergence as a divergence measure, and it is accurately and efficiently estimated by a method of direct density-ratio estimation. Through experiments on artificial and real-world datasets including human-activity sensing, speech, and Twitter messages, we demonstrate the usefulness of the proposed method.

stat.ML↗

arXiv2012

Density-Difference Estimation

We address the problem of estimating the difference between two probability densities. A naive approach is a two-step procedure of first estimating two densities separately and then computing their difference. However, such a two-step procedure does not necessarily work well because the first step is performed without regard to the second step and thus a small error incurred in the first stage can cause a big error in the second stage. In this paper, we propose a single-shot procedure for directly estimating the density difference without separately estimating two densities. We derive a non-parametric finite-sample error bound for the proposed single-shot density-difference estimator and show that it achieves the optimal convergence rate. The usefulness of the proposed method is also demonstrated experimentally.

cs.LG↗

arXiv2012

Super-sensitive Molecule-hugging Graphene Nanopores

Longitudinal resolution and lateral sensitivity are decisive characteristics that determine the suitability of a nanopore sensor for sequencing a strand of DNA as well as other important polymers. Previous modeling of DNA induced ionic current blockades in single atom thick graphene nanopores has shown these nanopores to have sufficient longitudinal resolution to distinguish individual nucleobases along the length of a DNA molecule. Here we experimentally focus on the sensitivity to small changes in DNA diameter that can be discerned with graphene nanopores. We show that remarkably large sensitivities (0.5 nA/A)are obtained when the nanopore is tailored to have a diameter close to that of the polymer of interest. Our results have been obtained with double-stranded DNA (dsDNA). Smaller graphene nanopores that can be tuned to the diameter of single-stranded DNA (ssDNA) for sequencing have only recently been demonstrated. Our results indicate that nanopore sensors based on such pores will provide excellent resolution and base discrimination, and will be robust competitors to protein nanopores that have recently demonstrated sequencing capability.

physics.bio-ph↗

arXiv2011

Generic Multiplicative Methods for Implementing Machine Learning Algorithms on MapReduce

In this paper we introduce a generic model for multiplicative algorithms which is suitable for the MapReduce parallel programming paradigm. We implement three typical machine learning algorithms to demonstrate how similarity comparison, gradient descent, power method and other classic learning techniques fit this model well. Two versions of large-scale matrix multiplication are discussed in this paper, and different methods are developed for both cases with regard to their unique computational characteristics and problem settings. In contrast to earlier research, we focus on fundamental linear algebra techniques that establish a generic approach for a range of algorithms, rather than specific ways of scaling up algorithms one at a time. Experiments show promising results when evaluated on both speedup and accuracy. Compared with a standard implementation with computational complexity $O(m^3)$ in the worst case, the large-scale matrix multiplication experiments prove our design is considerably more efficient and maintains a good speedup as the number of cores increases. Algorithm-specific experiments also produce encouraging results on runtime performance.

cs.DS↗

arXiv2004

Unbound Protein-Protein Docking Selections by the DFIRE-based Statistical Pair Potential

A newly developed statistical pair potential based on Distance-scaled Finite Ideal-gas REference (DFIRE) state is applied to unbound protein-protein docking structure selections. The performance of the DFIRE energy function is compared to those of the well-established ZDOCK energy scores and RosettaDock energy function using the comprehensive decoy sets generated by ZDOCK and RosettaDock. Despite significant difference in the functional forms and complexities of the three energy scores, the differences in overall performance for docking structure selections are small between DFIRE and ZDOCK2.3 and between DFIRE and RosettaDock. This result is remarkable considering that a single-term DFIRE energy function was originally designed for monomer proteins while multiple-term energy functions of ZDOCK and RosettaDock were specifically optimized for docking. This provides hope that the accuracy of the existing energy functions for docking can be improved.

q-bio.BM↗