The idea is to promote a "cover" for the final video. This is produced automatically, using almost all available sensors and presented to the user, so that the video can be viewed by a larger number of spectators. The cover frame, this time, is suggested from the 15 frames Video senses. The default causes always appear first frame detected, explains TechCrunch. Now, with the application of such patents, Facebook and Instagram will be able to choose the most interesting moments and suggest tags and whatever else needed to use to make the video more appealing. The patents in question were registered in October 2011, but only approved in April 2013. The technology that Facebook holds lets look at each video frame as if it were a photograph. The algorithms created then allow users to automatically identify, words, sights and brands. This type of information may appear specifically in each video, or just like being associated metadata. In the latter situation, the videos on a particular monument could be displayed to users who had already referred to like that place. The tagging system can become much more complex than it currently is, but with these technologies much of the complexity can be automated. Finally, it was also patented technology that enables Instagram hear what is being said in each video and capture the moment more interesting and presents it as cover frame. For example, when someone yells "Goal" during a video, the speech detection algorithms can run to catch the earlier time when the player delivers the ball. The idea behind these patents is that the video tags begin functioning as recognition that it is done in the photographs.
segunda-feira, 8 de julho de 2013
Facebook wants videos "smart" on Instagram
Facebook has registered patents that make a video may have geographical location, face recognition and even interpret the sound picked up by a microphone. The videos of Instagram can come to know and hear everything.
The idea is to promote a "cover" for the final video. This is produced automatically, using almost all available sensors and presented to the user, so that the video can be viewed by a larger number of spectators. The cover frame, this time, is suggested from the 15 frames Video senses. The default causes always appear first frame detected, explains TechCrunch. Now, with the application of such patents, Facebook and Instagram will be able to choose the most interesting moments and suggest tags and whatever else needed to use to make the video more appealing. The patents in question were registered in October 2011, but only approved in April 2013. The technology that Facebook holds lets look at each video frame as if it were a photograph. The algorithms created then allow users to automatically identify, words, sights and brands. This type of information may appear specifically in each video, or just like being associated metadata. In the latter situation, the videos on a particular monument could be displayed to users who had already referred to like that place. The tagging system can become much more complex than it currently is, but with these technologies much of the complexity can be automated. Finally, it was also patented technology that enables Instagram hear what is being said in each video and capture the moment more interesting and presents it as cover frame. For example, when someone yells "Goal" during a video, the speech detection algorithms can run to catch the earlier time when the player delivers the ball. The idea behind these patents is that the video tags begin functioning as recognition that it is done in the photographs.
The idea is to promote a "cover" for the final video. This is produced automatically, using almost all available sensors and presented to the user, so that the video can be viewed by a larger number of spectators. The cover frame, this time, is suggested from the 15 frames Video senses. The default causes always appear first frame detected, explains TechCrunch. Now, with the application of such patents, Facebook and Instagram will be able to choose the most interesting moments and suggest tags and whatever else needed to use to make the video more appealing. The patents in question were registered in October 2011, but only approved in April 2013. The technology that Facebook holds lets look at each video frame as if it were a photograph. The algorithms created then allow users to automatically identify, words, sights and brands. This type of information may appear specifically in each video, or just like being associated metadata. In the latter situation, the videos on a particular monument could be displayed to users who had already referred to like that place. The tagging system can become much more complex than it currently is, but with these technologies much of the complexity can be automated. Finally, it was also patented technology that enables Instagram hear what is being said in each video and capture the moment more interesting and presents it as cover frame. For example, when someone yells "Goal" during a video, the speech detection algorithms can run to catch the earlier time when the player delivers the ball. The idea behind these patents is that the video tags begin functioning as recognition that it is done in the photographs.
Assinar:
Postar comentários (Atom)
Nenhum comentário:
Postar um comentário