May 7, 2019

3039 words 15 mins read

Paper Group ANR 62

DLAU: A Scalable Deep Learning Accelerator Unit on FPGA. Dissecting a Social Botnet: Growth, Content and Influence in Twitter. Facial Surface Analysis using Iso-Geodesic Curves in Three Dimensional Face Recognition System. Memory Efficient Multi-Scale Line Detector Architecture for Retinal Blood Vessel Segmentation. A Hierarchical Pose-Based Approa …

DLAU: A Scalable Deep Learning Accelerator Unit on FPGA


Title	DLAU: A Scalable Deep Learning Accelerator Unit on FPGA
Authors	Chao Wang, Qi Yu, Lei Gong, Xi Li, Yuan Xie, Xuehai Zhou
Abstract	As the emerging field of machine learning, deep learning shows excellent ability in solving complex learning problems. However, the size of the networks becomes increasingly large scale due to the demands of the practical applications, which poses significant challenge to construct a high performance implementations of deep learning neural networks. In order to improve the performance as well to maintain the low power cost, in this paper we design DLAU, which is a scalable accelerator architecture for large-scale deep learning networks using FPGA as the hardware prototype. The DLAU accelerator employs three pipelined processing units to improve the throughput and utilizes tile techniques to explore locality for deep learning applications. Experimental results on the state-of-the-art Xilinx FPGA board demonstrate that the DLAU accelerator is able to achieve up to 36.1x speedup comparing to the Intel Core2 processors, with the power consumption at 234mW.
Tasks
Published	2016-05-23
URL	http://arxiv.org/abs/1605.06894v1
PDF	http://arxiv.org/pdf/1605.06894v1.pdf
PWC	https://paperswithcode.com/paper/dlau-a-scalable-deep-learning-accelerator
Repo
Framework


Title	Dissecting a Social Botnet: Growth, Content and Influence in Twitter
Authors	Norah Abokhodair, Daisy Yoo, David W. McDonald
Abstract	Social botnets have become an important phenomenon on social media. There are many ways in which social bots can disrupt or influence online discourse, such as, spam hashtags, scam twitter users, and astroturfing. In this paper we considered one specific social botnet in Twitter to understand how it grows over time, how the content of tweets by the social botnet differ from regular users in the same dataset, and lastly, how the social botnet may have influenced the relevant discussions. Our analysis is based on a qualitative coding for approximately 3000 tweets in Arabic and English from the Syrian social bot that was active for 35 weeks on Twitter before it was shutdown. We find that the growth, behavior and content of this particular botnet did not specifically align with common conceptions of botnets. Further we identify interesting aspects of the botnet that distinguish it from regular users.
Tasks
Published	2016-04-13
URL	http://arxiv.org/abs/1604.03627v1
PDF	http://arxiv.org/pdf/1604.03627v1.pdf
PWC	https://paperswithcode.com/paper/dissecting-a-social-botnet-growth-content-and
Repo
Framework

Facial Surface Analysis using Iso-Geodesic Curves in Three Dimensional Face Recognition System


Title	Facial Surface Analysis using Iso-Geodesic Curves in Three Dimensional Face Recognition System
Authors	Rachid Ahdid, El Mahdi Barrah, Said Safi, Bouzid Manaut
Abstract	In this paper, we present an automatic 3D face recognition system. This system is based on the representation of human faces surfaces as collections of Iso-Geodesic Curves (IGC) using 3D Fast Marching algorithm. To compare two facial surfaces, we compute a geodesic distance between a pair of facial curves using a Riemannian geometry. In the classifying step, we use: Neural Networks (NN), K-Nearest Neighbor (KNN) and Support Vector Machines (SVM). To test this method and evaluate its performance, a simulation series of experiments were performed on 3D Shape REtrieval Contest 2008 database (SHREC2008).
Tasks	3D Shape Retrieval, Face Recognition
Published	2016-08-31
URL	http://arxiv.org/abs/1608.08878v1
PDF	http://arxiv.org/pdf/1608.08878v1.pdf
PWC	https://paperswithcode.com/paper/facial-surface-analysis-using-iso-geodesic
Repo
Framework

Memory Efficient Multi-Scale Line Detector Architecture for Retinal Blood Vessel Segmentation


Title	Memory Efficient Multi-Scale Line Detector Architecture for Retinal Blood Vessel Segmentation
Authors	Hamza Bendaoudi, Farida Cheriet, J. M. Pierre Langlois
Abstract	This paper presents a memory efficient architecture that implements the Multi-Scale Line Detector (MSLD) algorithm for real-time retinal blood vessel detection in fundus images on a Zynq FPGA. This implementation benefits from the FPGA parallelism to drastically reduce the memory requirements of the MSLD from two images to a few values. The architecture is optimized in terms of resource utilization by reusing the computations and optimizing the bit-width. The throughput is increased by designing fully pipelined functional units. The architecture is capable of achieving a comparable accuracy to its software implementation but 70x faster for low resolution images. For high resolution images, it achieves an acceleration by a factor of 323x.
Tasks
Published	2016-12-06
URL	http://arxiv.org/abs/1612.09524v1
PDF	http://arxiv.org/pdf/1612.09524v1.pdf
PWC	https://paperswithcode.com/paper/memory-efficient-multi-scale-line-detector
Repo
Framework

A Hierarchical Pose-Based Approach to Complex Action Understanding Using Dictionaries of Actionlets and Motion Poselets


Title	A Hierarchical Pose-Based Approach to Complex Action Understanding Using Dictionaries of Actionlets and Motion Poselets
Authors	Ivan Lillo, Juan Carlos Niebles, Alvaro Soto
Abstract	In this paper, we introduce a new hierarchical model for human action recognition using body joint locations. Our model can categorize complex actions in videos, and perform spatio-temporal annotations of the atomic actions that compose the complex action being performed.That is, for each atomic action, the model generates temporal action annotations by estimating its starting and ending times, as well as, spatial annotations by inferring the human body parts that are involved in executing the action. our model includes three key novel properties: (i) it can be trained with no spatial supervision, as it can automatically discover active body parts from temporal action annotations only; (ii) it jointly learns flexible representations for motion poselets and actionlets that encode the visual variability of body parts and atomic actions; (iii) a mechanism to discard idle or non-informative body parts which increases its robustness to common pose estimation errors. We evaluate the performance of our method using multiple action recognition benchmarks. Our model consistently outperforms baselines and state-of-the-art action recognition methods.
Tasks	Pose Estimation, Temporal Action Localization
Published	2016-06-15
URL	http://arxiv.org/abs/1606.04992v1
PDF	http://arxiv.org/pdf/1606.04992v1.pdf
PWC	https://paperswithcode.com/paper/a-hierarchical-pose-based-approach-to-complex
Repo
Framework

Network Morphism


Title	Network Morphism
Authors	Tao Wei, Changhu Wang, Yong Rui, Chang Wen Chen
Abstract	We present in this paper a systematic study on how to morph a well-trained neural network to a new one so that its network function can be completely preserved. We define this as \emph{network morphism} in this research. After morphing a parent network, the child network is expected to inherit the knowledge from its parent network and also has the potential to continue growing into a more powerful one with much shortened training time. The first requirement for this network morphism is its ability to handle diverse morphing types of networks, including changes of depth, width, kernel size, and even subnet. To meet this requirement, we first introduce the network morphism equations, and then develop novel morphing algorithms for all these morphing types for both classic and convolutional neural networks. The second requirement for this network morphism is its ability to deal with non-linearity in a network. We propose a family of parametric-activation functions to facilitate the morphing of any continuous non-linear activation neurons. Experimental results on benchmark datasets and typical neural networks demonstrate the effectiveness of the proposed network morphism scheme.
Tasks
Published	2016-03-05
URL	http://arxiv.org/abs/1603.01670v2
PDF	http://arxiv.org/pdf/1603.01670v2.pdf
PWC	https://paperswithcode.com/paper/network-morphism
Repo
Framework

Graph-Community Detection for Cross-Document Topic Segment Relationship Identification


Title	Graph-Community Detection for Cross-Document Topic Segment Relationship Identification
Authors	Pedro Mota, Maxine Eskenazi, Luisa Coheur
Abstract	In this paper we propose a graph-community detection approach to identify cross-document relationships at the topic segment level. Given a set of related documents, we automatically find these relationships by clustering segments with similar content (topics). In this context, we study how different weighting mechanisms influence the discovery of word communities that relate to the different topics found in the documents. Finally, we test different mapping functions to assign topic segments to word communities, determining which topic segments are considered equivalent. By performing this task it is possible to enable efficient multi-document browsing, since when a user finds relevant content in one document we can provide access to similar topics in other documents. We deploy our approach in two different scenarios. One is an educational scenario where equivalence relationships between learning materials need to be found. The other consists of a series of dialogs in a social context where students discuss commonplace topics. Results show that our proposed approach better discovered equivalence relationships in learning material documents and obtained close results in the social speech domain, where the best performing approach was a clustering technique.
Tasks	Community Detection
Published	2016-06-13
URL	http://arxiv.org/abs/1606.04081v1
PDF	http://arxiv.org/pdf/1606.04081v1.pdf
PWC	https://paperswithcode.com/paper/graph-community-detection-for-cross-document
Repo
Framework

SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound


Title	SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand Ultrasound
Authors	Christian F. Baumgartner, Konstantinos Kamnitsas, Jacqueline Matthew, Tara P. Fletcher, Sandra Smith, Lisa M. Koch, Bernhard Kainz, Daniel Rueckert
Abstract	Identifying and interpreting fetal standard scan planes during 2D ultrasound mid-pregnancy examinations are highly complex tasks which require years of training. Apart from guiding the probe to the correct location, it can be equally difficult for a non-expert to identify relevant structures within the image. Automatic image processing can provide tools to help experienced as well as inexperienced operators with these tasks. In this paper, we propose a novel method based on convolutional neural networks which can automatically detect 13 fetal standard views in freehand 2D ultrasound data as well as provide a localisation of the fetal structures via a bounding box. An important contribution is that the network learns to localise the target anatomy using weak supervision based on image-level labels only. The network architecture is designed to operate in real-time while providing optimal output for the localisation task. We present results for real-time annotation, retrospective frame retrieval from saved videos, and localisation on a very large and challenging dataset consisting of images and video recordings of full clinical anomaly screenings. We found that the proposed method achieved an average F1-score of 0.798 in a realistic classification experiment modelling real-time detection, and obtained a 90.09% accuracy for retrospective frame retrieval. Moreover, an accuracy of 77.8% was achieved on the localisation task.
Tasks
Published	2016-12-16
URL	http://arxiv.org/abs/1612.05601v2
PDF	http://arxiv.org/pdf/1612.05601v2.pdf
PWC	https://paperswithcode.com/paper/sononet-real-time-detection-and-localisation
Repo
Framework

Universal Correspondence Network


Title	Universal Correspondence Network
Authors	Christopher B. Choy, JunYoung Gwak, Silvio Savarese, Manmohan Chandraker
Abstract	We present a deep learning framework for accurate visual correspondences and demonstrate its effectiveness for both geometric and semantic matching, spanning across rigid motions to intra-class shape or appearance variations. In contrast to previous CNN-based approaches that optimize a surrogate patch similarity objective, we use deep metric learning to directly learn a feature space that preserves either geometric or semantic similarity. Our fully convolutional architecture, along with a novel correspondence contrastive loss allows faster training by effective reuse of computations, accurate gradient computation through the use of thousands of examples per image pair and faster testing with $O(n)$ feed forward passes for $n$ keypoints, instead of $O(n^2)$ for typical patch similarity methods. We propose a convolutional spatial transformer to mimic patch normalization in traditional features like SIFT, which is shown to dramatically boost accuracy for semantic correspondences across intra-class shape variations. Extensive experiments on KITTI, PASCAL, and CUB-2011 datasets demonstrate the significant advantages of our features over prior works that use either hand-constructed or learned features.
Tasks	Metric Learning, Semantic Similarity, Semantic Textual Similarity
Published	2016-06-11
URL	http://arxiv.org/abs/1606.03558v3
PDF	http://arxiv.org/pdf/1606.03558v3.pdf
PWC	https://paperswithcode.com/paper/universal-correspondence-network
Repo
Framework

Search by Ideal Candidates: Next Generation of Talent Search at LinkedIn


Title	Search by Ideal Candidates: Next Generation of Talent Search at LinkedIn
Authors	Viet Ha-Thuc, Ye Xu, Satya Pradeep Kanduri, Xianren Wu, Vijay Dialani, Yan Yan, Abhishek Gupta, Shakti Sinha
Abstract	One key challenge in talent search is how to translate complex criteria of a hiring position into a search query. This typically requires deep knowledge on which skills are typically needed for the position, what are their alternatives, which companies are likely to have such candidates, etc. However, listing examples of suitable candidates for a given position is a relatively easy job. Therefore, in order to help searchers overcome this challenge, we design a next generation of talent search paradigm at LinkedIn: Search by Ideal Candidates. This new system only needs the searcher to input one or several examples of suitable candidates for the position. The system will generate a query based on the input candidates and then retrieve and rank results based on the query as well as the input candidates. The query is also shown to the searcher to make the system transparent and to allow the searcher to interact with it. As the searcher modifies the initial query and makes it deviate from the ideal candidates, the search ranking function dynamically adjusts an refreshes the ranking results balancing between the roles of query and ideal candidates. As of writing this paper, the new system is being launched to our customers.
Tasks
Published	2016-02-26
URL	http://arxiv.org/abs/1602.08186v1
PDF	http://arxiv.org/pdf/1602.08186v1.pdf
PWC	https://paperswithcode.com/paper/search-by-ideal-candidates-next-generation-of
Repo
Framework

Do We Need Binary Features for 3D Reconstruction?


Title	Do We Need Binary Features for 3D Reconstruction?
Authors	Bin Fan, Qingqun Kong, Wei Sui, Zhiheng Wang, Xinchao Wang, Shiming Xiang, Chunhong Pan, Pascal Fua
Abstract	Binary features have been incrementally popular in the past few years due to their low memory footprints and the efficient computation of Hamming distance between binary descriptors. They have been shown with promising results on some real time applications, e.g., SLAM, where the matching operations are relative few. However, in computer vision, there are many applications such as 3D reconstruction requiring lots of matching operations between local features. Therefore, a natural question is that is the binary feature still a promising solution to this kind of applications? To get the answer, this paper conducts a comparative study of binary features and their matching methods on the context of 3D reconstruction in a recently proposed large scale mutliview stereo dataset. Our evaluations reveal that not all binary features are capable of this task. Most of them are inferior to the classical SIFT based method in terms of reconstruction accuracy and completeness with a not significant better computational performance.
Tasks	3D Reconstruction
Published	2016-02-14
URL	http://arxiv.org/abs/1602.04502v1
PDF	http://arxiv.org/pdf/1602.04502v1.pdf
PWC	https://paperswithcode.com/paper/do-we-need-binary-features-for-3d
Repo
Framework

A convolutional approach to reflection symmetry


Title	A convolutional approach to reflection symmetry
Authors	Marcelo Cicconet, Vighnesh Birodkar, Mads Lund, Michael Werman, Davi Geiger
Abstract	We present a convolutional approach to reflection symmetry detection in 2D. Our model, built on the products of complex-valued wavelet convolutions, simplifies previous edge-based pairwise methods. Being parameter-centered, as opposed to feature-centered, it has certain computational advantages when the object sizes are known a priori, as demonstrated in an ellipse detection application. The method outperforms the best-performing algorithm on the CVPR 2013 Symmetry Detection Competition Database in the single-symmetry case. Code and a new database for 2D symmetry detection is available.
Tasks
Published	2016-09-17
URL	http://arxiv.org/abs/1609.05257v1
PDF	http://arxiv.org/pdf/1609.05257v1.pdf
PWC	https://paperswithcode.com/paper/a-convolutional-approach-to-reflection
Repo
Framework

Fast Algorithm of High-resolution Microwave Imaging Using the Non-parametric Generalized Reflectivity Model


Title	Fast Algorithm of High-resolution Microwave Imaging Using the Non-parametric Generalized Reflectivity Model
Authors	Long Gang Wang, Lianlin Li, Tie Jun Cui
Abstract	This paper presents an efficient algorithm of high-resolution microwave imaging based on the concept of generalized reflectivity. The contribution made in this paper is two-fold. We introduce the concept of non-parametric generalized reflectivity (GR, for short) as a function of operational frequencies and view angles, etc. The GR extends the conventional Born-based imaging model, i.e., single-scattering model, into that accounting for more realistic interaction between the electromagnetic wavefield and imaged scene. Afterwards, the GR-based microwave imaging is formulated in the convex of sparsity-regularized optimization. Typically, the sparsity-regularized optimization requires the implementation of iterative strategy, which is computationally expensive, especially for large-scale problems. To break this bottleneck, we convert the imaging problem into the problem of physics-driven image processing by introducing a dual transformation. Moreover, this image processing is performed over overlapping patches, which can be efficiently solved in the parallel or distributed manner. In this way, the proposed high-resolution imaging methodology could be applicable to large-scale microwave imaging problems. Selected simulation results are provided to demonstrate the state-of-art performance of proposed methodology.
Tasks
Published	2016-09-12
URL	http://arxiv.org/abs/1611.03341v1
PDF	http://arxiv.org/pdf/1611.03341v1.pdf
PWC	https://paperswithcode.com/paper/fast-algorithm-of-high-resolution-microwave
Repo
Framework

Memory Visualization for Gated Recurrent Neural Networks in Speech Recognition


Title	Memory Visualization for Gated Recurrent Neural Networks in Speech Recognition
Authors	Zhiyuan Tang, Ying Shi, Dong Wang, Yang Feng, Shiyue Zhang
Abstract	Recurrent neural networks (RNNs) have shown clear superiority in sequence modeling, particularly the ones with gated units, such as long short-term memory (LSTM) and gated recurrent unit (GRU). However, the dynamic properties behind the remarkable performance remain unclear in many applications, e.g., automatic speech recognition (ASR). This paper employs visualization techniques to study the behavior of LSTM and GRU when performing speech recognition tasks. Our experiments show some interesting patterns in the gated memory, and some of them have inspired simple yet effective modifications on the network structure. We report two of such modifications: (1) lazy cell update in LSTM, and (2) shortcut connections for residual learning. Both modifications lead to more comprehensible and powerful networks.
Tasks	Speech Recognition
Published	2016-09-28
URL	http://arxiv.org/abs/1609.08789v3
PDF	http://arxiv.org/pdf/1609.08789v3.pdf
PWC	https://paperswithcode.com/paper/memory-visualization-for-gated-recurrent
Repo
Framework

Incremental Minimax Optimization based Fuzzy Clustering for Large Multi-view Data


Title	Incremental Minimax Optimization based Fuzzy Clustering for Large Multi-view Data
Authors	Yangtao Wang, Lihui Chen, Xiaoli Li
Abstract	Incremental clustering approaches have been proposed for handling large data when given data set is too large to be stored. The key idea of these approaches is to find representatives to represent each cluster in each data chunk and final data analysis is carried out based on those identified representatives from all the chunks. However, most of the incremental approaches are used for single view data. As large multi-view data generated from multiple sources becomes prevalent nowadays, there is a need for incremental clustering approaches to handle both large and multi-view data. In this paper we propose a new incremental clustering approach called incremental minimax optimization based fuzzy clustering (IminimaxFCM) to handle large multi-view data. In IminimaxFCM, representatives with multiple views are identified to represent each cluster by integrating multiple complementary views using minimax optimization. The detailed problem formulation, updating rules derivation, and the in-depth analysis of the proposed IminimaxFCM are provided. Experimental studies on several real world multi-view data sets have been conducted. We observed that IminimaxFCM outperforms related incremental fuzzy clustering in terms of clustering accuracy, demonstrating the great potential of IminimaxFCM for large multi-view data analysis.
Tasks
Published	2016-08-25
URL	http://arxiv.org/abs/1608.07001v1
PDF	http://arxiv.org/pdf/1608.07001v1.pdf
PWC	https://paperswithcode.com/paper/incremental-minimax-optimization-based-fuzzy
Repo
Framework