Systems and methods for evaluating online videos
Systems and methods for evaluating online videos. One method includes receiving a URL; determining a URL type; detecting whether the URL includes one or more videos; determining at least one of a size of the video, a position of the video on a web page of the web page URL, whether the video is set to autoplay, and whether the video is set to mute; computing a score based on one or more of the size of the video, the position of the video on the web page of the web page URL, whether the video is set to autoplay, and whether the video is set to mute; obtaining at least two frames of at least part of the video, wherein each frame is obtained at one or more predetermined intervals during playback of the video; and classifying each detected video based on the at least two frames.
1. A computer-implemented method for evaluating online videos, the method comprising:
receiving, at one or more servers, a plurality of URLs to be reviewed in relation to a web domain;
detecting, by the one or more servers, for any URL determined to be a web page URL of the web domain, whether a web page of the URL includes one or more videos;
if a video is detected on a web page of a web page URL, optionally determining, by the one or more servers, for any video detected on a web page of a web page URL of the web domain, at least one of a size of the video and a position of the video on a web page of the web page URL;
if a video is detected on a web page of a web page URL, optionally computing, by the one or more servers, for any video detected on a web page URL of the web domain, a viewability score based on the determined size of the video and the determined position of the video on the web page of the web page URL;
identifying, by the one or more servers, for any URL determined to be a video URL, a video to be classified by the one or more servers;
classifying, by the one or more servers, each video detected on a web page of a web page URL and each video of a video URL of the web domain, based on analysis of at least two frames of at least part of the video, wherein each frame is obtained by a frame grabber at one or more predetermined intervals during playback of the video;
analyzing, by the one or more servers, the classification result of each video using at least one of statistical techniques and heuristic techniques;
computing and storing, by the one or more servers, a rate at which video content of the web domain is changed based on a comparison of videos detected on the web domain against videos previously detected on the web domain, wherein the web domain is visited automatically and at a predefined frequency to perform the computing;
generating, by the one or more servers, an assay to perform analysis of classified videos for the web domain based on the classifying, wherein the assay automatically visits the web domain to perform analysis; and
determining, by the one or more servers, a frequency for repeating the assay for the web domain based on the computed and stored rate at which video content of the web domain is changed, and automatically repeating the assay based on the determined frequency.
2. The method of claim 1 , further comprising:
receiving, for a web page URL, a web page,
wherein obtaining at least two frames of at least part of the video comprises:
obtaining, by the one or more servers, for each video detected on the web page URL, at least two screenshots of at least part of the video on the web page, wherein each screenshot is obtained at one or more predetermined intervals during playback of the video; and
obtaining, by the one or more servers, for each video detected on a web page URL, each frame of the at least two frames from a screenshot of the at least two screenshots based on the size of the video and the position of the video on the web page, wherein each frame of the at least two frames is obtained from a different screenshot of the at least two screenshots.
3. The method of claim 1 , further comprising:
generating an assay of classified videos for the web domain based on the classifying; and
setting or receiving an advertising bid or an advertising rate for the web domain based on the generated assay for the web domain or the computed rate of the web domain.
4. The method of claim 1 , wherein classifying each detected video based on the at least two frames uses at least one of computer vision techniques and crowdsourcing techniques.
5. The method of claim 1 , further comprising:
filtering, by the one or more servers, the videos detected on the web pages of the web domain based on the classification result of each video.
6. The method of claim 1 , further comprising:
storing, by the one or more servers, at least one of the viewability score and the classification result computed for each video in at least one database.
7. The method of claim 1 , wherein the one or more predetermined intervals includes intervals of about 0.1 seconds to about 10 seconds.
8. A system for evaluating online videos, the system comprising:
a memory device storing instructions for evaluating online videos; and
a processor configured to execute the instructions to perform a method of:
receiving, at one or more servers, a plurality of URLs to be reviewed in relation to a web domain;
detecting, by the one or more servers, for any URL determined to be a web page URL of the web domain, whether a web page of the URL includes one or more videos;
if a video is detected on a web page of a web page URL, optionally determining, by the one or more servers, for any video detected on a web page of a web page URL of the web domain, at least one of a size of the video, and a position of the video on a web page of the web page URL;
if a video is detected on a web page of a web page URL, optionally computing, by the one or more servers, for any video detected on a web page URL of the web domain, a viewability score based on the determined size of the video and the determined position of the video on the web page of the web page URL;
identifying, by the one or more servers, for any URL determined to be a video URL, a video to be classified by the one or more servers;
classifying, by the one or more servers, each video detected on a web page of a web page URL and each video of a video URL of the web domain, based on analysis of at least two frames of at least part of the video, wherein each frame is obtained by a frame grabber at one or more predetermined intervals during playback of the video;
analyzing, by the one or more servers, the classification result of each video using at least one of statistical techniques and heuristic techniques;
computing and storing, by the one or more servers, a rate at which video content of the web domain is changed based on a comparison of videos detected on the web domain against videos previously detected on the web domain, wherein the web domain is visited automatically and at a predefined frequency to perform the computing;
generating, by the one or more servers, an assay to perform analysis of classified videos for the web domain based on the classifying, wherein the assay automatically visits the web domain to perform the analysis; and
determining, by the one or more servers, a frequency for repeating the assay for the web domain based on the computed and stored rate at which video content of the web domain is changed, and automatically repeating the assay based on the determined frequency.
9. The system of claim 8 , wherein the processor is further configured to execute the instructions to perform:
receiving, for a web page URL, a web page,
wherein obtaining at least two frames of at least part of the video comprises:
obtaining, by the one or more servers, for each video detected on the web page URL, at least two screenshots of at least part of the video on the web page, wherein each screenshot is obtained at one or more predetermined intervals during playback of the video; and
obtaining, by the one or more servers, for each video detected on a web page URL, each frame of the at least two frames from a screenshot of the at least two screenshots based on the size of the video and the position of the video on the web page, wherein each frame of the at least two frames is obtained from a different screenshot of the at least two screenshots.
10. The system of claim 8 , further comprising:
generating an assay of classified videos for the web domain based on the classifying; and
setting or receiving an advertising bid or an advertising rate for the web domain based on the generated assay for the web domain or the computed rate of the web domain.
11. The system of claim 8 , wherein classifying each detected video based on the at least two frames uses at least one of computer vision techniques and crowdsourcing techniques.
12. The system of claim 8 , wherein the processor is further configured to execute the instructions to perform:
filtering, by the one or more servers, the videos detected on the web pages of the web domain based on the classification result of each video.
13. The system of claim 8 , wherein the processor is further configured to execute the instructions to perform:
storing, by the one or more servers, at least one of the viewability score and the classification result computed for each video in at least one database.
14. The system of claim 8 , wherein the one or more predetermined intervals includes intervals of about 0.1 seconds to about 10 seconds.