Advances in Multimedia

Retracted: Enhanced P2P Services Providing Multimedia Content

This paper proposes and evaluates the LFrWF, a novel lifting-based architecture to compute the discrete wavelet transform (DWT) of images using the fractional wavelet filter (FrWF). In order to reduce the memory requirement of the proposed architecture, only one image line is read into a buffer at a time. Aside from an LFrWF version with multipliers, i.e., the LFr<span class="nowrap"><svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M1" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><path id="g190-88" d="M956 650H736V622C814 613 818 602 804 542C789 471 733 275 691 141H689C628 314 570 476 514 641H485C434 478 377 312 323 142H321C277 284 238 419 199 554C184 605 186 613 259 622V650H16V622C77 615 92 606 117 524C172 346 228 170 281 -11H312C363 151 419 312 475 488H477C528 343 597 150 652 -11H683C765 269 826 458 848 522C878 605 887 613 956 622V650Z"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><path id="g190-71" d="M493 503C489 551 484 614 483 650H43V622C120 616 128 611 128 525V126C128 40 120 34 40 28V0H312V28C221 34 213 40 213 126V307H316C407 307 412 296 424 227H453V420H424C412 355 407 346 316 346H213V584C213 613 216 616 246 616H322C398 616 419 607 436 579C449 559 455 539 464 499L493 503Z"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><path id="g190-110" d="M797 0V26C739 32 732 36 732 103V296C732 394 682 449 605 449C576 449 550 437 529 423C504 407 475 389 446 366C425 418 382 449 334 449C303 449 279 437 253 421C222 403 201 385 180 371V452C135 432 85 419 41 411V388C99 379 102 374 102 310V103C102 38 93 32 27 26V0H238V26C189 32 180 38 180 103V338C210 363 250 390 289 390C351 390 377 348 377 275V103C377 37 368 32 306 26V0H520V26C465 32 456 38 456 101V296C456 314 455 326 453 338C491 369 529 390 565 390C628 390 653 345 653 274V107C653 36 642 32 583 26V0H797Z"/></g></svg>,</span> we develop a multiplier-less LFrWF version, i.e., the LFr<span class="nowrap"><svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M2" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><path id="g190-109" d="M238 0V26C174 32 166 38 166 104V712C132 700 70 683 18 677V653C81 647 87 645 87 577V104C87 38 78 32 15 26V0H238Z"/></g></svg>,</span> which reduces the critical path delay (CPD) to the delay <svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.29108pt" id="M3" height="12.2532pt" version="1.1" viewBox="-0.0498162 -8.96212 12.4956 12.2532" width="12.4956pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><path id="g113-85" d="M620 675H597C578 656 570 650 541 650H144C112 650 104 653 94 675H72C59 618 42 552 23 493L53 491C71 534 88 564 105 585C124 608 144 615 238 615H290L197 121C182 40 174 34 88 28L82 0H361L367 28C275 34 266 38 281 121L374 615H441C522 615 543 608 553 583C562 560 566 531 565 493L597 494C603 551 612 629 620 675Z"/></g><g transform="matrix(.0091,0,0,-0.0091,7.176,3.132)"><path id="g50-98" d="M490 97L476 124C442 96 405 70 398 70C392 70 390 78 396 114C419 243 448 379 463 432L457 436C446 436 431 439 418 442C393 447 368 451 343 451C281 451 204 418 155 381C74 320 24 206 24 107C24 23 59 -12 88 -12C118 -12 155 5 191 34C236 70 290 122 328 177H330L312 84C296 0 311 -12 331 -12C355 -12 425 24 490 97ZM374 387C371 367 360 299 347 264C323 202 187 53 142 53C128 53 113 73 113 120C113 224 157 332 221 380C241 395 274 403 303 403C330 403 360 395 374 387Z"/></g></svg> of an adder. The proposed LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M4" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g></svg> and LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M5" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><use xlink:href="#g190-109"/></g></svg> architectures are compared in terms of the required adders, multipliers, memory, and critical path delay with state-of-the-art DWT architectures. Moreover, the proposed LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M6" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g></svg> and LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M7" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><use xlink:href="#g190-109"/></g></svg> architectures, along with the state-of-the-art FrWF architectures (with multipliers (Fr<span class="nowrap"><svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M8" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g></svg>)</span> and without multipliers (Fr<span class="nowrap"><svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M9" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><use xlink:href="#g190-109"/></g></svg>))</span> are compared through implementation on the same FPGA board. The LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M10" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g></svg> requires 22% less look-up tables (LUT), 34% less flip-flops (FF), and 50% less compute cycles (CC) and consumes 65% less energy than the Fr<span class="nowrap"><svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M11" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g></svg>.</span> Also, the proposed LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M12" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><use xlink:href="#g190-109"/></g></svg> architecture requires 50% less CC and consumes 43% less energy than the Fr<span class="nowrap"><svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M13" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><use xlink:href="#g190-109"/></g></svg>.</span> Thus, the proposed LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M14" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 27.6517 11.7065" width="27.6517pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g></svg> and LFr<svg xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" style="vertical-align:-3.18152pt" id="M15" height="11.7065pt" version="1.1" viewBox="-0.0498162 -8.52498 30.0083 11.7065" width="30.0083pt"><g transform="matrix(.013,0,0,-0.013,0,0)"><use xlink:href="#g190-88"/></g><g transform="matrix(.013,0,0,-0.013,12.623,0)"><use xlink:href="#g190-71"/></g><g transform="matrix(.0091,0,0,-0.0091,19.573,3.132)"><use xlink:href="#g190-110"/></g><g transform="matrix(.0091,0,0,-0.0091,27.072,3.132)"><use xlink:href="#g190-109"/></g></svg> architectures appear suitable for computing the DWT of images on wearable sensors.

Mohd Tausif

Ekram Khan

Mohd Hasan

Martin Reisslein

Lifting-Based Fractional Wavelet Filter: Energy-Efficient DWT Architecture for Low-Cost Wearable Sensors

This study attempts to predict stock index prices using multivariate time series analysis. The study&#x2019;s motivation is based on the notion that datasets of stock index prices involve weak periodic patterns, long-term and short-term information, for which traditional approaches and current neural networks such as Autoregressive models and Support Vector Machine (SVM) may fail. This study applied Temporal Pattern Attention and Long-Short-Term Memory (TPA-LSTM) for prediction to overcome the issue. The results show that stock index prices prediction through the TPA-LSTM algorithm could achieve better prediction performance over traditional deep neural networks, such as recurrent neural network (RNN), convolutional neural network (CNN), and long and short-term time series network (LSTNet).

Xiaolu Wei

Binbin Lei

Hongbing Ouyang

Qiufeng Wu

Stock Index Prices Prediction via Temporal Pattern Attention and Long-Short-Term Memory

recognit human emot from facial express affect distort pictori qualiti facial pose which often ignor tradit video emot recognit method other hand context inform can also provid differ degre extra clue which can further improv recognit accuraci paper first build video dataset seven categori human emot name human emot video heiv with heiv dataset train contextawar attent network caan recogn human emot the network consist two subnetwork process both face context inform featur from facial express context clue fuse repres emot video frame which pass through attent network generat emot score then emot featur all frame aggreg accord emot score experiment result show our propos method effect heiv dataset

Xiaodong Liu

Miao Wang

Context-Aware Attention Network for Human Emotion Recognition in Video

paper symmetri detect algorithm threedimension point cloud model base weight princip compon analysi pca propos the propos algorithm work follow first use point element area initi weight weight pca perform plane select initi symmetri plane iter method use adjust approxim symmetri plane step step make tend perfect symmetri plane domin symmetri plane each iter first updat weight each point base distanc metric use new weight perform weight pca determin new symmetri plane current plane symmetri close enough plane symmetri previous iter number iter exceed given threshold iter termin after iter termin plane symmetri last iter taken domin symmetri plane model shown experiment result propos algorithm can find domin symmetri plane symmetr model also work well nonperfect symmetr model

Chen He

Lei Wang

Yonghui Zhang

Chunmeng Wang

Dominant Symmetry Plane Detection for Point-Based 3D Models

digit imag watermark attract research area sinc protect multimedia data from unauthor access for design effici robust digit imag watermark system tradeoff among impercept robust capac secur must maintain various studi regard concern have been perform ensur requir hybrid differ domain spatial transform domain paper have present analyt studi exist hybrid digit imag watermark method first have given standard framework design hybrid method ensur basic design requir watermark various applic after brief literatur review compar analyz complex sever exist hybrid method tabular form the limit applic method also highlight final summar challeng exist method conclud studi give futur research direct

Mahbuba Begum

Mohammad Shorif Uddin

Analysis of Digital Image Watermarking Techniques through Hybrid Methods

while reconstruct object increas use today simplif point cloud howev becom substanti phase process reconstruct this due huge amount dens point cloud produc scan devic paper new approach propos simplifi point cloud base nearest neighbor cluster algorithm initial point cloud divid cluster use mean algorithm then entropi estim perform each cluster remov one have minim entropi paper matlab use carri out simul perform our method testifi test dataset numer experi demonstr effect propos simplif method point cloud

Abdelaaziz Mahdaoui

El Hassan Sbai

3D Point Cloud Simplification Based on <i>k</i>-Nearest Neighbor and Clustering

paper studi challeng imagetovideo retriev which use queri imag search relev frame from larg collect video novel framework base convolut neural network cnns propos perform largescal video retriev low storag cost high search effici our framework consist keyfram extract algorithm featur aggreg strategi specif keyfram extract algorithm take advantag cluster idea redund inform remov video data storag cost great reduc the featur aggreg strategi adopt averag pool encod deep local convolut featur follow coarsetofin retriev which allow rapid retriev largescal video databas the result from extens experi two public avail dataset demonstr propos method achiev superior effici well accuraci over other stateoftheart visual search method

Chen Zhang

Bin Hu

Yucong Suo

Zhiqiang Zou

Yimu Ji

Large-Scale Video Retrieval via Deep Local Convolutional Features

grass one most import resourc ecosystem sustain develop human be howev studi focus grass identif which were tradit implement expert low effici precis cannot meet requir modern grassland manag studi propos cubic interpol lbp cilbp dbn wavelet grass identif base leaf imag lowfrequ compon leaf imag decompos dbn wavelet use input cilbp more subtl textur extract the novelti propos method cilbp can better describ textur featur from lowfrequ subimag compar origin bilinear lbp the effect identif accuraci propos method grass leaf demonstr experiment result

Fan Han

Xue Qiao

Yubao Ma

Weihong Yan

Xinyu Wang

Xin Pan

Grass Leaf Identification Using dbN Wavelet and CILBP

this paper present new multisupervis coupl metric learn mscml method lowresolut face imag match while coupl metric learn has achiev good perform degrad face recognit most exist coupl metric learn method onli adopt categori label supervis which easili lead chang distribut sampl coupl space and accuraci degrad imag match serious influenc chang address problem propos mscml method train linear nonlinear metric model respect which can project differ resolut face pair same latent featur space under which distanc each posit pair reduc each negat pair enlarg work defin novel multisupervis object function which consist main object function auxiliari object function the supervis inform main object function categori label which play major supervisori role the supervis inform auxiliari object function distribut relationship sampl which play auxiliari supervisori role under supervis categori label distribut inform learn model can better deal intraclass multimod problem featur obtain coupl space more easili match correct experiment result three differ face dataset valid efficaci propos method

Guofeng Zou

Guixia Fu

Xiang Peng

A Novel Multisupervised Coupled Metric Learning for Low-Resolution Face Matching

pedestrian detect classif increas interest intellig transport system its among challeng issu can find limit tini occlud appear larg variat human pose clutter background complex environ fact partial occlus handl import case detect pedestrian order avoid accid between pedestrian vehicl sinc difficult detect when pedestrian sudden cross road solv partial occlus problem pyramid partbas model ppm propos obtain more accur predict base major vote confid score visibl part cascad pyramid structur the experiment result propos scheme achiev 9625 accuraci inria dataset 81 accuraci psu princ songkla univers dataset our propos model can appli realworld environ classifi occlud part pedestrian various inform part represent each pyramid layer

M. Thu

N. Suvonvorn

Pyramidal Part-Based Model for Partial Occlusion Handling in Pedestrian Classification

About this journal

<p><em>Advances in Multimedia</em> publishes research on the technologies associated with multimedia systems, including computer-media integration for digital information processing, storage, transmission, and representation.</p>


Journal profile

Meet the editorial board

<p><em>Advances in Multimedia </em>maintains an Editorial Board of practicing researchers from around the world, to ensure manuscripts are handled by editors who are experts in the field of study.</p>


Editor spotlight

Submit your proposal

Do you think there is an emerging area of research that really needs to be highlighted? Or an existing research area that has been overlooked or would benefit from deeper investigation? Raise the profile of a research area by leading a Special Issue.

Advances in Multimedia