Inspiration
Most of the time when watching a video we're only concerned with a certain part of the video and not the whole. We may be trying to show our friends a funny scene in a show or find the parts of a soccer game where Cristiano Ronaldo scores a goal.
What it does
Our platform receives areas of interest from the user in the form of tags, and searches multiple videos in our database and finds the timestamps within the video where the specified tags are present.
How we built it
This project uses google cloud video intelligence to extract transcripts from the videos. We extract tags and keywords from these transcripts using the rake open source api. These tags are stored along with the corresponding timestamps and video data in the google cloud datastore. When tags are searched, we query the database for a response and the best matches selected. The top match is displayed, and the other matches are listed beneath the video.
Accomplishments that we're proud of
We are proud of the accuracy of our platform in terms of how well the database responses match the queried tags
What's next for tl;dw
we could integrate object detection into our project to better the accuracy of our system
Log in or sign up for Devpost to join the conversation.