完整後設資料紀錄
DC 欄位語言
dc.contributor.author周宏春zh_TW
dc.contributor.author蔡文錦zh_TW
dc.contributor.authorChou, Hung-Chunen_US
dc.contributor.authorTsai, Wen-Jiinen_US
dc.date.accessioned2018-01-24T07:41:00Z-
dc.date.available2018-01-24T07:41:00Z-
dc.date.issued2015en_US
dc.identifier.urihttp://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070256003en_US
dc.identifier.urihttp://hdl.handle.net/11536/141450-
dc.description.abstract本論文利用旋積類神經網路學習具有鑑別能力的特徵描述子以用於影像檢索。近年來由於旋積類神經網路的特徵學習方法在物件識別應用上有相當不錯的表現,許多研究嘗試將此用於影像檢索上。然而,直接將事先訓練好用於影像分類的權重參數應用於影像檢索,其效能可能不是最好的。為了解決這個問題,我們提出藉由對比損失函數改寫旋積類神經網路的權重參數,使得生成的特徵值適合用於影像檢索應用。我們將在標準測試的影像檢索資料庫上進行研究,實驗結果顯示本論文所提出之架構優於目前最先進之各方法。zh_TW
dc.description.abstractThe thesis aims to learn discriminative features for image retrieval tasks based on using deep convolutional neural networks (CNN). Motivated by the great success of CNN in recognition tasks, one may be tempted to simply adopt the output of CNN for retrieval. However, CNN pre-trained model for classification tasks may not optimized for retrieval tasks. To address this issue, the CNN’s weight parameters are specifically adapted by a contrastive loss function to suit retrieval tasks. Extensive experiments conducted on typical retrieval datasets confirm the superiority of the proposed scheme over the state-of-the-art methods.en_US
dc.language.isozh_TWen_US
dc.subject影像檢索zh_TW
dc.subjectImage Retrievalen_US
dc.title針對影像搜尋應用中學習具鑑別能力之旋積類神經網路描述子zh_TW
dc.titleDiscriminatively-learned CNN Features for Image Retrievalen_US
dc.typeThesisen_US
dc.contributor.department資訊科學與工程研究所zh_TW
顯示於類別:畢業論文