首頁瀏覽人次: 641,721 次 (Since 2010/07/14) 會員登入
聯盟單位  |  虛擬講堂  |  網路直播  |  其他演講  |  排行榜  |  留言板  |  知識擂台  |  回首頁
首頁 > 虛擬講堂 > Video Super-Resolution Using Convolutional Neural Network 快速搜尋演講 進階搜尋
  Video Super-Resolution Using Convolutional Neural Network 
★chrome瀏覽器無法播放Silverlight解決辦法
0:00 / 0:00

講者:國立成功大學電機工程學系-Chien-Wei Lin學生
日期:2017/10/25
性質:演講
類別:物質科學
語言:中文
長度:00:11:56
觀看:116
推薦:0
摘要:
Convolutional neural network (CNN) has been widely applied to super-resolution tasks. However, the existing super-resolution algorithms wi...
Convolutional neural network (CNN) has been widely applied to super-resolution tasks. However, the existing super-resolution algorithms with CNN need pre-processing steps which require massive computation. In this paper, we propose an architecture that learns both spatial and temporal information of low-resolution video. Consecutive frames are used as input to our video super-resolution CNN. Unlike existing approaches, only a single convolutional layer, named correlational layer, is used to replace the motion compensation step. Experimental results show that the proposed correlational layer is able to learn the motion information and outperform the network with motion compensated frames by 0.19 dB in PSNR. Our method can averagely outperformVSRnet by 0.39 dB in PSNR for videos upscaled with factor 3.

現在位置:演講摘要詳細內容
推薦  (0)
推薦至Plurk
提供:TANET台灣網際網路研討會-TANET2017

轉寄  

推薦者:
電子郵件地址: (如欲轉寄多人,請以 ; 分隔email)
留言給收件者:

回報問題  

問題說明:




提升公共給水震損評估與管材耐...
講者:劉季宇
觀看:619
腦與行為的關係
講者:洪蘭
觀看:22,264
從化學小分子看大自然現象
講者:張海舟
觀看:580
共振與波動的魔法—淺談光子晶...
講者:欒丕綱
觀看:218
大機器、小粒子—邁向物質核心...
講者:郭家銘
觀看:208
從科技趨勢預言反脆弱的黑科技...
講者:黃爾文
觀看:174

現在位置:學習公約