In general, watch and view the whole video is often time-consuming which involves displaying the video film linearly. In this paper, we propose a webpage-based and video summarization-based learning platform (WVSUM). The learning platform provides not only video lecture preview but also combines image, text, and video information as learning materials. The input videos can be automatically transformed into online lectures without human annotators through the subtitle extraction techniques. By means of the extracted caption words, the designed text summarization method is applied to generate the summary. Our summarization approach can generate variant length of video summary by setting up the time constraint. To validate the effectiveness, we compare with the existing fast forwards (one kind of video surrogate interface) and evaluate users' comprehension to video content in the limited viewing time. Thirty undergraduate students were invited to examine the video learning platforms. The experimental results showed that our WVSUM had better effect than fast forwards on comprehension to videos. In terms of system usage and satisfaction, our WVSUM achieved significantly better scores than the classic fast forwards surrogate.