Capturing and storing elements from a video presentation for later retrieval in response to queries
View Patent ↗Embodiments of the present invention enable the extraction, classification, storage, and supplementation of presentation video. A media system receives a video signal carrying presentation video. The media system processes the video signal and generates images for slides of the presentation. The media system then extracts text from the images and uses the text and other characteristics to classify the images and store them in a database. Additionally, the system enables viewers of the presentation to provide feedback on the presentation, which can be used to supplement the presentation.
1. A method for capturing video presentations, the method comprising:
detecting, performed by one or more microprocessors, when a video presentation has begun by analyzing an incoming video stream;
determining whether a presentation element in the video presentation is static or dynamic in response to detecting that the video presentation has begun;
generating an image from the presentation element when the presentation element is static;
generating a video clip from the presentation element when the presentation element is dynamic;
extracting textual content of the presentation element from the image;
associating a plurality of weights with a plurality of text of the textual content of the presentation element based on types of the plurality of text such that a first weight is associated with a first text of the plurality of text that is not a title of the presentation element and a second weight is associated with a second text of the plurality of text of the presentation element, the second weight being greater than the first weight and associated with the second text in response to determining that the second text is a title of the presentation element, the first text and the second text being weighted textual content;
receiving supplemental content corresponding to the image or the video clip;
storing the weighted textual content and the supplemental content in a searchable index;
detecting when the video presentation has stopped by analyzing the incoming video stream; and
stopping the generation of the video clip and the image in response to detecting that the video presentation has stopped,
wherein in response to the extracted textual content associated with the image, stored in the searchable index, matching a search query:
audio content is extracted from the generated video clip, and
the audio content associated with the image is played and returned in response to the search query.
2. The method of claim 1 , wherein the generated image from the presentation element comprises a slide image bitmap.
3. The method of claim 1 , wherein when at least one of the extracted textual content associated with the image, stored in the searchable index, matches a search query, the image is displayed and returned in response to the search query.
4. The method of claim 1 , wherein the generated video clip is extracted from the incoming video stream.
5. The method of claim 1 , wherein the audio content is extracted from the incoming video stream.
6. The method of claim 1 , wherein the method is performed by the one or more microprocessors.
7. The method of claim 1 , wherein the type of types of the plurality of text comprise section heading text, title text and body text.
8. The method of claim 1 , wherein the receiving the supplemental content corresponding to the image or the video clip comprises receiving a user comment or a user question.
9. The method of claim 1 , further comprising associating the video presentation with a plurality of other video presentations in response to determining that the video presentation and the other video presentations are presented on a same day.
10. The method of claim 1 , further comprising associating the presentation element with an overlay requesting display over the presentation element and received over a network connection while the presentation element is displayed.
11. A non-transitory computer readable storage medium storing a program causing a computer to execute a process for capturing video presentations, the process comprising:
detecting, performed by one or more microprocessors, when a video presentation has begun by analyzing an incoming video stream;
determining whether a presentation element in the video presentation is static or dynamic in response to detecting that the video presentation has begun;
generating an image from the presentation element when the presentation element is static;
generating a video clip from the presentation element when the presentation element is dynamic;
extracting textual content of the presentation element from the image;
associating a plurality of weights with a plurality of text of the textual content of the presentation element based on types of the plurality of text such that a first weight is associated with a first text of the plurality of text that is not a title of the presentation element and a second weight is associated with a second text of the plurality of text of the presentation element, the second weight being greater than the first weight and associated with the second text in response to determining that the second text is a title of the presentation element, the first text and the second text being weighted textual content;
receiving supplemental content corresponding to the image or the video clip;
storing the weighted textual content and the supplemental content in a searchable index;
detecting when the video presentation has stopped by analyzing the incoming video stream; and
stopping the generation of the video clip and the image in response to detecting that the video presentation has stopped,
wherein in response to the extracted textual content associated with the image, stored in the searchable index, matching a search query:
audio content is extracted from the generated video clip, and
the audio content associated with the image is played and returned in response to the search query.