arXiv subjects

Songyang Zhang

Publications and source records attributed to Songyang Zhang.

At least 145 records · Page 8Linked to original sources

arXiv2020

From Spectrum Wavelet to Vertex Propagation: Graph Convolutional Networks Based on Taylor Approximation

Graph convolutional networks (GCN) have been recently utilized to extract the underlying structures of datasets with some labeled data and high-dimensional features. Existing GCNs mostly rely on a first-order Chebyshev approximation of graph wavelet-kernels. Such a generic propagation model does not always suit the various datasets and their features. This work revisits the fundamentals of graph wavelet and explores the utility of signal propagation in the vertex domain to approximate the spectral wavelet-kernels. We first derive the conditions for representing the graph wavelet-kernels via vertex propagation. We next propose alternative propagation models for GCN layers based on Taylor expansions. We further analyze the choices of detailed graph representations for TGCNs. Experiments on citation networks, multimedia datasets and synthetic graphs demonstrate the advantage of Taylor-based GCN (TGCN) in the node classification problems over the traditional GCN methods.

eess.SP↗

arXiv2020

Content-based Analysis of the Cultural Differences between TikTok and Douyin

Short-form video social media shifts away from the traditional media paradigm by telling the audience a dynamic story to attract their attention. In particular, different combinations of everyday objects can be employed to represent a unique scene that is both interesting and understandable. Offered by the same company, TikTok and Douyin are popular examples of such new media that has become popular in recent years, while being tailored for different markets (e.g. the United States and China). The hypothesis that they express cultural differences together with media fashion and social idiosyncrasy is the primary target of our research. To that end, we first employ the Faster Regional Convolutional Neural Network (Faster R-CNN) pre-trained with the Microsoft Common Objects in COntext (MS-COCO) dataset to perform object detection. Based on a suite of objects detected from videos, we perform statistical analysis including label statistics, label similarity, and label-person distribution. We further use the Two-Stream Inflated 3D ConvNet (I3D) pre-trained with the Kinetics dataset to categorize and analyze human actions. By comparing the distributional results of TikTok and Douyin, we uncover a wealth of similarity and contrast between the two closely related video social media platforms along the content dimensions of object quantity, object categories, and human action categories.

cs.CV↗

arXiv2020

Transformer with Bidirectional Decoder for Speech Recognition

Attention-based models have made tremendous progress on end-to-end automatic speech recognition(ASR) recently. However, the conventional transformer-based approaches usually generate the sequence results token by token from left to right, leaving the right-to-left contexts unexploited. In this work, we introduce a bidirectional speech transformer to utilize the different directional contexts simultaneously. Specifically, the outputs of our proposed transformer include a left-to-right target, and a right-to-left target. In inference stage, we use the introduced bidirectional beam search method, which can not only generate left-to-right candidates but also generate right-to-left candidates, and determine the best hypothesis by the score. To demonstrate our proposed speech transformer with a bidirectional decoder(STBD), we conduct extensive experiments on the AISHELL-1 dataset. The results of experiments show that STBD achieves a 3.6\% relative CER reduction(CERR) over the unidirectional speech transformer baseline. Besides, the strongest model in this paper called STBD-Big can achieve 6.64\% CER on the test set, without language model rescoring and any extra data augmentation strategies.

eess.AS↗

arXiv2020

Point Cloud Segmentation based on Hypergraph Spectral Clustering

Hypergraph spectral analysis has emerged as an effective tool processing complex data structures in data analysis. The surface of a three-dimensional (3D) point cloud and the multilateral relationship among their points can be naturally captured by the high-dimensional hyperedges. This work investigates the power of hypergraph spectral analysis in unsupervised segmentation of 3D point clouds. We estimate and order the hypergraph spectrum from observed point cloud coordinates. By trimming the redundancy from the estimated hypergraph spectral space based on spectral component strengths, we develop a clustering-based segmentation method. We apply the proposed method to various point clouds, and analyze their respective spectral properties. Our experimental results demonstrate the effectiveness and efficiency of the proposed segmentation method.

eess.SP↗

arXiv2020

LearningCC: An online learning approach for congestion control

Recently, much effort has been devoted by researchers from both academia and industry to develop novel congestion control methods. LearningCC is presented in this letter, in which the congestion control problem is solved by reinforce learning approach. Instead of adjusting the congestion window with fixed policy, there are serval options for an endpoint to choose. To predict the best option is a hard task. Each option is mapped as an arm of a bandit machine. The endpoint can learn to determine the optimal choice through trial and error method. Experiments are performed on ns3 platform to verify the effectiveness of LearningCC by comparing with other benchmark algorithms. Results indicate it can achieve lower transmission delay than loss based algorithms. Especially, we found LearningCC makes significant improvement in link suffering from random loss.

cs.NI↗

arXiv2020

Global Image Sentiment Transfer

Transferring the sentiment of an image is an unexplored research topic in the area of computer vision. This work proposes a novel framework consisting of a reference image retrieval step and a global sentiment transfer step to transfer sentiments of images according to a given sentiment tag. The proposed image retrieval algorithm is based on the SSIM index. The retrieved reference images by the proposed algorithm are more content-related against the algorithm based on the perceptual loss. Therefore can lead to a better image sentiment transfer result. In addition, we propose a global sentiment transfer step, which employs an optimization algorithm to iteratively transfer sentiment of images based on feature maps produced by the Densenet121 architecture. The proposed sentiment transfer algorithm can transfer the sentiment of images while ensuring the content structure of the input image intact. The qualitative and quantitative experiments demonstrate that the proposed sentiment transfer framework outperforms existing artistic and photorealistic style transfer algorithms in making reliable sentiment transfer results with rich and exact details.

cs.CV↗

arXiv2020

Introducing Hypergraph Signal Processing: Theoretical Foundation and Practical Applications

Signal processing over graphs has recently attracted significant attentions for dealing with structured data. Normal graphs, however, only model pairwise relationships between nodes and are not effective in representing and capturing some high-order relationships of data samples, which are common in many applications such as Internet of Things (IoT). In this work, we propose a new framework of hypergraph signal processing (HGSP) based on tensor representation to generalize the traditional graph signal processing (GSP) to tackle high-order interactions. We introduce the core concepts of HGSP and define the hypergraph Fourier space. We then study the spectrum properties of hypergraph Fourier transform and explain its connection to mainstream digital signal processing. We derive the novel hypergraph sampling theory and present the fundamentals of hypergraph filter design based on the tensor framework. We present HGSP-based methods for several signal processing and data analysis applications. Our experimental results demonstrate significant performance improvement using our HGSP framework over some traditional signal processing solutions.

eess.SP↗

arXiv2020

An Online Learning Based Path Selection for Multipath Video Telephony Service in Overlay

Even real time video telephony services have been pervasively applied, providing satisfactory quality of experience to users is still a challenge task especially in wireless networks. Multipath transmission is a promising solution to improve video quality by aggregating bandwidth. In existing multipath transmission solutions, sender concurrently splits traffic on default routing paths and has no flexibility to select paths. When default paths fall into severe congestion and the available bandwidth decreases, sender has to decrease video quality by reducing resolution or encoding bitrate. Deploying relay servers in current infrastructure to form overlay network provides path diversity. An online learning approach based on multi-armed bandits is applied for path selection to harvest maximum profit. Further, a congestion control algorithm adapted from BBR is implemented to probe bandwidth and to avoid link congestion. To maintain throughput stability and fairness, a smaller probe up gain value is used and the cycle length in bandwidth probe phase is randomized. To reduce delay, the inflight packets will be reduced to match with the estimated bandwidth delay product in the probe down phase. Experiments are conducted to verify the effectiveness the proposed solution to improve throughput and quality in video communication service.

cs.NI↗

arXiv2020

Mi YouTube es Su YouTube? Analyzing the Cultures using YouTube Thumbnails of Popular Videos

YouTube, a world-famous video sharing website, maintains a list of the top trending videos on the platform. Due to its huge amount of users, it enables researchers to understand people's preference by analyzing the trending videos. Trending videos vary from country to country. By analyzing such differences and changes, we can tell how users' preferences differ over locations. Previous work focuses on analyzing such culture preferences from videos' metadata, while the culture information hidden within the visual content has not been discovered. In this study, we explore culture preferences among countries using the thumbnails of YouTube trending videos. We first process the thumbnail images of the videos using object detectors. The collected object information is then used for various statistical analysis. In particular, we examine the data from three perspectives: geographical locations, video genres and users' reactions. Experimental results indicate that the users from similar cultures shares interests in watching similar videos on YouTube. Our study demonstrates that discovering the culture preference through the thumbnails can be an effective mechanism for video social media analysis.

cs.SI↗

arXiv2020

Hypergraph Spectral Analysis and Processing in 3D Point Cloud

Along with increasingly popular virtual reality applications, the three-dimensional (3D) point cloud has become a fundamental data structure to characterize 3D objects and surroundings. To process 3D point clouds efficiently, a suitable model for the underlying structure and outlier noises is always critical. In this work, we propose a hypergraph-based new point cloud model that is amenable to efficient analysis and processing. We introduce tensor-based methods to estimate hypergraph spectrum components and frequency coefficients of point clouds in both ideal and noisy settings. We establish an analytical connection between hypergraph frequencies and structural features. We further evaluate the efficacy of hypergraph spectrum estimation in two common point cloud applications of sampling and denoising for which also we elaborate specific hypergraph filter design and spectral properties. The empirical performance demonstrates the strength of hypergraph signal processing as a tool in 3D point clouds and the underlying properties.

eess.SP↗

arXiv2019

Learning Sparse 2D Temporal Adjacent Networks for Temporal Action Localization

In this report, we introduce the Winner method for HACS Temporal Action Localization Challenge 2019. Temporal action localization is challenging since a target proposal may be related to several other candidate proposals in an untrimmed video. Existing methods cannot tackle this challenge well since temporal proposals are considered individually and their temporal dependencies are neglected. To address this issue, we propose sparse 2D temporal adjacent networks to model the temporal relationship between candidate proposals. This method is built upon the recent proposed 2D-TAN approach. The sampling strategy in 2D-TAN introduces the unbalanced context problem, where short proposals can perceive more context than long proposals. Therefore, we further propose a Sparse 2D Temporal Adjacent Network (S-2D-TAN). It is capable of involving more context information for long proposals and further learning discriminative features from them. By combining our S-2D-TAN with a simple action classifier, our method achieves a mAP of 23.49 on the test set, which win the first place in the HACS challenge.

cs.CV↗

arXiv2019

An Evaluation of BBR and its variants

The congestion control algorithm bring such importance that it avoids the network link into severe congestion and guarantees network normal operation. Since The loss based algorithms introduce high transmission delay, to design new algorithm simultaneously achieving high throughout and low buffer occupation is a new working direction. The bottleneck bandwidth and round trip time (BBR) belongs such kind, and it has drawn much attention since its release. There are other algorithms modified from BBR to gain better performance. And the implementation of BBR v2.0 is released recently. We implement a framework to compare the performance of these algorithms in simulated environment.

cs.NI↗

arXiv2019

Dynamic Context Correspondence Network for Semantic Alignment

Establishing semantic correspondence is a core problem in computer vision and remains challenging due to large intra-class variations and lack of annotated data. In this paper, we aim to incorporate global semantic context in a flexible manner to overcome the limitations of prior work that relies on local semantic representations. To this end, we first propose a context-aware semantic representation that incorporates spatial layout for robust matching against local ambiguities. We then develop a novel dynamic fusion strategy based on attention mechanism to weave the advantages of both local and context features by integrating semantic cues from multiple scales. We instantiate our strategy by designing an end-to-end learnable deep network, named as Dynamic Context Correspondence Network (DCCNet). To train the network, we adopt a multi-auxiliary task loss to improve the efficiency of our weakly-supervised learning procedure. Our approach achieves superior or competitive performance over previous methods on several challenging datasets, including PF-Pascal, PF-Willow, and TSS, demonstrating its effectiveness and generality.

cs.CV↗

arXiv2019

Exploiting Temporal Relationships in Video Moment Localization with Natural Language

We address the problem of video moment localization with natural language, i.e. localizing a video segment described by a natural language sentence. While most prior work focuses on grounding the query as a whole, temporal dependencies and reasoning between events within the text are not fully considered. In this paper, we propose a novel Temporal Compositional Modular Network (TCMN) where a tree attention network first automatically decomposes a sentence into three descriptions with respect to the main event, context event and temporal signal. Two modules are then utilized to measure the visual similarity and location similarity between each segment and the decomposed descriptions. Moreover, since the main event and context event may rely on different modalities (RGB or optical flow), we use late fusion to form an ensemble of four models, where each model is independently trained by one combination of the visual input. Experiments show that our model outperforms the state-of-the-art methods on the TEMPO dataset.

cs.MM↗

arXiv2019

LatentGNN: Learning Efficient Non-local Relations for Visual Recognition

Capturing long-range dependencies in feature representations is crucial for many visual recognition tasks. Despite recent successes of deep convolutional networks, it remains challenging to model non-local context relations between visual features. A promising strategy is to model the feature context by a fully-connected graph neural network (GNN), which augments traditional convolutional features with an estimated non-local context representation. However, most GNN-based approaches require computing a dense graph affinity matrix and hence have difficulty in scaling up to tackle complex real-world visual problems. In this work, we propose an efficient and yet flexible non-local relation representation based on a novel class of graph neural networks. Our key idea is to introduce a latent space to reduce the complexity of graph, which allows us to use a low-rank representation for the graph affinity matrix and to achieve a linear complexity in computation. Extensive experimental evaluations on three major visual recognition tasks show that our method outperforms the prior works with a large margin while maintaining a low computation cost.

cs.CV↗

arXiv2019

An Optimized BBR for Multipath Real Time Video Streaming

The multipath transmission scheme can work as an effective way to provide better quality of experiments to end users. Two key research points in the multipath real time video transmission context are congestion control and packet scheduling. As Utility maximization theory shows, to provide better satisfaction to end users is to provide higher throughput and lower transmission delay. The congestion control is responsible to converge to the maximum available bandwidth and avoid leading the network into congestion. A delay response BBR (Delay-BBR) algorithm optimized for real time video transmission is proposed, and the main idea is to reduce sending rate when the link delay has exceeded certain threshold to let the intermediate routers drain the occupied buffer. It can achieve better transmission delay and lower packet loss rate compared with QUIC-BBR and WebRTC-BBR by experiment. And a packet scheduling algorithm induced from Utility maximization theory works on top of the congestion control algorithm is tested and achieves lower frame delivery delay further compared with benchmark algorithms.

cs.NI↗

arXiv2018

Shared Bottleneck Detecction Based on Trend Line Regression for Multipath Transmission

The current deployed multipath congestion control algorithms couple all the subflows together to avoid bandwidth occupation aggressiveness if the subflows of multipath transmission protocol share common bottleneck with single path TCP. The coupled congestion control algorithms can guarantee well fairness property in common bottleneck but result in rate increase conservativeness in none-sharing bottleneck situation. Thus, the throughput of multipath session can be further improved when combing with effective shared bottleneck detection mechanism. This paper proposes a delay trend line regression method to detect if flows share common bottleneck. Deduced from TCP fluid model, the packet round trip delay signal shows linear increase property during the queue building up process of the narrowest link and the delay trend line slopes of two flows are in close proximity if they traverse the same bottleneck link. The proposed method is implemented on multipath QUIC golang codebase and extensive simulations are performed to validate its effectiveness in detecting out flows traversing common bottleneck. If the subflows are detected out via a common bottleneck, the sender would perform coupled congestion control algorithm and perform congestion control seperately on flow level in none sharing bottleneck case. Results show a multipath session with two subflows can obtain 74\% gain on average in throughput compared with single path connection when Linked Increases Algorithm (LIA) is in combination with trend line regession shared bottlenck detection algorithm in none shared bottleneck, and show well fairness property in common bottleneck scenarios.

cs.NI↗

arXiv2018

Congestion Control for RTP Media: a Comparison on Simulated Environment

To develop low latency congestion control algorithm for real time taffic has been gained attention recently. RTP Media Congestion Avoidance Techniques (RMCAT) working group was initiated for standard defination. There are three algorithms under this group, Network Assisted Dynamic Adaptation (NADA) proposed by Cisco, Google Congestion Control (GCC) proposed by Google and Self-Clocked Rate Adaptation for Multimedia(SCReAM) proposed by Ericsson. This paper compares and analyses the performance of these algorithms on simulation environment. Results show GCC has well fairness property and performs well in lossy link but slow convergence in dynamic link, NADA stabilizes its rate quickly but suffers "late-comer" effect, SCReAM has the lowest queue occupation but also lower link capacity utilization.

cs.NI↗