国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁(yè)> 外文OA文獻(xiàn) >Shallow and deep convolutional networks for saliency prediction
【2h】

Shallow and deep convolutional networks for saliency prediction

機(jī)譯：用于顯著性預(yù)測(cè)的淺層和深層卷積網(wǎng)絡(luò)

代理獲取
本網(wǎng)站僅為用戶提供外文OA文獻(xiàn)查詢和代理獲取服務(wù)，本網(wǎng)站沒(méi)有原文。下單后我們將采用程序或人工為您竭誠(chéng)獲取高質(zhì)量的原文，但由于OA文獻(xiàn)來(lái)源多樣且變更頻繁，仍可能出現(xiàn)獲取不到、文獻(xiàn)不完整或與標(biāo)題不符等情況，如果獲取不到我們將提供退款服務(wù)。請(qǐng)知悉。
  • 摘要
  • 著錄項(xiàng)
  • 相似文獻(xiàn)
  • 相關(guān)主題

摘要

The prediction of salient areas in images has been traditionally addressed with hand-crafted features based on neuroscience principles. This paper, however, addresses the problem with a completely data-driven approach by training a convolutional neural network (convnet). The learning process is formulated as a minimization of a loss function that measures the Euclidean distance of the predicted saliency map with the provided ground truth. The recent publication of large datasets of saliency prediction has provided enough data to train end-to-end architectures that are both fast and accurate. Two designs are proposed: a shallow convnet trained from scratch, and a another deeper solution whose first three layers are adapted from another network trained for classification. To the authors knowledge, these are the first end-to-end CNNs trained and tested for the purpose of saliency prediction.
機(jī)譯：傳統(tǒng)上，基于神經(jīng)科學(xué)原理通過(guò)手工制作的功能解決了圖像顯著區(qū)域的預(yù)測(cè)問(wèn)題。但是，本文通過(guò)訓(xùn)練卷積神經(jīng)網(wǎng)絡(luò)（convnet），以完全數(shù)據(jù)驅(qū)動(dòng)的方法解決了該問(wèn)題。學(xué)習(xí)過(guò)程被表述為損失函數(shù)的最小值，該損失函數(shù)使用提供的地面真實(shí)性來(lái)測(cè)量預(yù)測(cè)顯著性圖的歐幾里得距離。最近發(fā)布的顯著性預(yù)測(cè)大型數(shù)據(jù)集提供了足夠的數(shù)據(jù)來(lái)訓(xùn)練快速而準(zhǔn)確的端到端架構(gòu)。提出了兩種設(shè)計(jì)：從頭開(kāi)始訓(xùn)練的淺卷積網(wǎng)絡(luò)，以及另一種更深層次的解決方案，其前三層改編自另一種經(jīng)過(guò)訓(xùn)練的分類網(wǎng)絡(luò)。據(jù)作者所知，這些是為顯著性預(yù)測(cè)目的而經(jīng)過(guò)培訓(xùn)和測(cè)試的首批端到端CNN。

著錄項(xiàng)

相似文獻(xiàn)

  • 外文文獻(xiàn)
  • 中文文獻(xiàn)
  • 專利
代理獲取

客服郵箱：kefu@zhangqiaokeyan.com

京公網(wǎng)安備：11010802029741號(hào) ICP備案號(hào)：京ICP備15016152號(hào)-6 六維聯(lián)合信息科技 (北京) 有限公司?版權(quán)所有

  • 客服微信

  • 服務(wù)號(hào)