Interactive video generation
A method includes receiving an input from a user, the input including a selection of an object, where the object is contained in a video, identifying, in response to the input, the object in the video, determining an attribute of the object associated with a profile of the user, where the profile contains a preference of the user regarding at least one object to be viewed, and determining, in response to the attribute of the object associated with the profile, a target segment in the video, where the target segment includes the object.
1. A method comprising:
receiving, by a device, an input from a user, the input comprising a selection of an object contained in a video played on the device, the selection is made via a touch gesture performed by a user's finger on a touchscreen of the device;
determining a frame of the video comprising an image displayed at a time the input from the user is received;
determining a position of the input of the user on the frame;
identifying the object in the determined frame by applying image recognition techniques, the identifying of the object from the frame is based on the determined position and an object position list, the object position list comprising information about positions of objects in the video;
comparing attributes of the object with user preferences defined in a user profile, the attributes comprising object characteristics, an object type, an object target customers, object price, and an object manufacturer, and the user preferences comprising one or more characteristics of the object previously selected by the user;
in response to at least one attribute of the object matching at least one user preference defined in the user profile, identifying segments from the video which display the object, each segment comprising a duration and one or more frames;
calculating a total sum of the duration of the identified segments;
comparing the total sum to a predetermined time interval; and
based on the calculated sum being equal to the predetermined time interval, combining the identified segments to provide a video tailored to the user preferences specified in the user profile.
2. The method of claim 1 , wherein the image recognition techniques comprise edge matching, divide-and-conquer search, greyscale matching, and gradient matching.
3. The method of claim 1 , further comprising:
generating the attributes for the object based on the user profile.
4. The method of claim 1 , further comprising:
selecting, in response to the calculated sum exceeding the predetermined time interval, a portion of each identified segment based on the user profile; and
combining the portion of each identified segment.
5. The method of claim 1 , further comprising:
determining, in response to the calculated sum being less than the predetermined time interval, weights for the identified segments based on attributes of the identified segments;
weighting the duration of the identified segments; and
combining the identified segments.
6. A method comprising:
receiving, by a device, a user input comprising a touch gesture performed by a user's finger on a touchscreen of the device, wherein an object is selected among one or more objects contained in a video played by the device via the touch gesture, the object comprising a product, a person, or a facility of interest to a user of the device;
determining a frame of the video comprising an image displayed at a time the user's finger contacts the touchscreen;
determining a first set of coordinates corresponding to a position of the user's finger on the frame;
obtaining an object position list comprising a frame id associated with a second set of coordinates for each object in the one or more objects, wherein each second set of coordinates corresponds to a location range associated with each of the one or more objects in the video;
based on the first set of coordinates being located within the second set of coordinates of at least one object within the one or more objects, recognizing the at least one object as the object selected by the user's finger;
comparing attributes of the object with user preferences defined by a user profile, the attributes comprising object characteristics, an object type, an object target customers, object price, and an object manufacturer, and the user preferences comprising one or more characteristics of the object previously selected by the user;
in response to at least one attribute of the object matching at least one user preference defined by the user profile, identifying segments from the video which display the object, each segment comprising a duration and one or more frames;
calculating a total sum of the duration of the identified segments;
comparing the total sum to a predetermined time interval; and
based on the calculated sum being equal to the predetermined time interval, combining the identified segments to provide a video tailored to the user preferences specified in the user profile.
7. The method of claim 6 , further comprising:
generating the attributes for the object based on the user profile.
8. The method of claim 6 , further comprising:
selecting, in response to the calculated sum exceeding the predetermined time interval, a portion of each identified segment based on the user profile; and
combining the portion of each identified segment.
9. The method of claim 6 , further comprising:
determining, in response to the calculated sum being less than the predetermined time interval, weights for the identified segments based on attributes of the identified segments;
weighting the duration of the identified segments; and
combining the identified segments.