{"id":5909,"date":"2014-10-23T04:41:13","date_gmt":"2014-10-23T11:41:13","guid":{"rendered":"http:\/\/palblog.fxpal.com\/?p=5909"},"modified":"2014-10-20T15:16:31","modified_gmt":"2014-10-20T22:16:31","slug":"video-text-retouch","status":"publish","type":"post","link":"https:\/\/blog.fxpal.net\/?p=5909","title":{"rendered":"video text retouch"},"content":{"rendered":"

Several of us just returned from ACM UIST 2014<\/a> where we presented some new work as part of the\u00a0cemint<\/a> project. \u00a0One vision of the cemint project is to build\u00a0applications for multimedia content\u00a0manipulation and reuse that are as powerful as their analogues for text content. \u00a0We are\u00a0working towards this goal\u00a0by exploiting two key tools. \u00a0First, we want to use real-time content analysis to expose useful structure within multimedia content. \u00a0Given some\u00a0decomposition of the content, which can be spatial, temporal, or even semantic, we then allow users to interact with these sub-units or segments via direct manipulation. \u00a0Last year, we began exploring these ideas in our work on content-based\u00a0video copy and paste<\/a>.<\/p>\n

As another\u00a0embodiment of these ideas, we demonstrated video text retouch at UIST last week. \u00a0Our browser-based system\u00a0performs real-time\u00a0text detection on streamed video frames to locate both\u00a0words and lines. \u00a0When a user clicks on a frame, a live cursor appears next to the nearest word. \u00a0At this point, users can alter text directly using the keyboard. \u00a0When they do so, a video overlay is created to capture and display their edits.<\/p>\n