| DC 欄位 |
值 |
語言 |
| DC.contributor | 資訊工程學系在職專班 | zh_TW |
| DC.creator | 彭信穎 | zh_TW |
| DC.creator | Hsin-Ying Peng | en_US |
| dc.date.accessioned | 2021-7-27T07:39:07Z | |
| dc.date.available | 2021-7-27T07:39:07Z | |
| dc.date.issued | 2021 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=108552018 | |
| dc.contributor.department | 資訊工程學系在職專班 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 目前臉部交換的應用大多都是要使用好幾萬張的使用者人臉照片以及被
交換的人臉照片，並且需要有一些程式基礎相關，且閱讀大量的文件說明，來
去做人臉交換應用的使用，過程中還會遇到各種環境問題，以及未知的錯誤，
本研究提出以網頁前後端加上 AWS 雲服務，並搭配人臉交換的應用程式，此架
構使用 Generator 與 Discriminator，並且主要目的是能夠用少量張數，來讓神經
網路學習圖像的風格到內容圖像的物體上，此程式只需要運行於後端 server，
在後端 server 將會搭配 nginx 以達負載均衡，使用者僅需提供一張使用者照片，
以及想要替換的目標的影片，並且能夠選擇目標影片中想要被交換的男主角或
女主角，透過網頁前端上傳後，再選擇欲被交換的對象，後端接收使用者的訊
息後，待後端處理完成，即可以取得最後成功換臉的影片。 | zh_TW |
| dc.description.abstract | Most of the current face exchange applications use tens of thousands of user face photos and the exchanged face photos, and need to have some program basics knowledge, and read a lot of document descriptions to make face exchange applications. In the process of use, various environmental problems and unknown errors will be encountered. This research proposes to add AWS cloud services to the front and back ends of the web page, and to match the application of face exchange. This architecture uses Generator and Discriminator, and the main purpose is a small number of images can be used to allow the neural network to learn the style of the image on the object of the content image. This program only needs to run on the back- end server. The back-end server will be matched with nginx to achieve load balancing. The user only needs to provide a user photo, and the video of the target you want to replace, and you can select the male or female protagonist in the target film that you want to be exchanged. After uploading it through the front-end of the web page, you can select the object to be exchanged. Back-end After receiving the user′s message, after the back-end processing is completed, the final video of the successful face change can be obtained. | en_US |
| DC.subject | 生成對抗式網路 | zh_TW |
| DC.subject | 臉部置換 | zh_TW |
| DC.subject | Generative Adversarial Network | en_US |
| DC.subject | Face Swap | en_US |
| DC.title | 影片指定對象臉部置換系統 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Face Replacement System For Designated Subjects Of The Video | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |