Video recommendation system and method thereof

A video recommendation method applied in a video recommendation system, the method comprising:

• receiving feedback information of videos from multiple users, wherein the videos comprise a plurality of video segments, and each video segment comprises a plurality of frames;
• calculating at least one object feature of the plurality of video segments corresponding to the feedback information;
• calculating the popularity of the at least one object feature of the plurality of video segments;
• and
• recommending at least one recommended video segment from the plurality of video segments according to the popularity of the at least one object feature,
• wherein the at least one object feature comprises at least one high level feature and at least one low level feature of the plurality of video segments, and
• wherein the high level feature indicates object characteristics of the video segments or the frames close to explanation by human, and the low level features are close to the substantial properties of objects.

A video recommendation system, comprising:

• an interaction device configured to receive feedback information of a video from a user, wherein the video comprises a plurality of video segments, and each video segment comprises a plurality of frames;
• a feature calculating module, implemented by a processor, configured to calculate object features of at least a first video segment corresponding to the feedback information in the plurality of video segments, wherein the object features comprise at least one high level feature and at least one low level feature of the at least one first video segment, and the high level feature indicates object characteristics of the video segments or the frames close to explanation by human, and the low level features are close to the substantial properties of objects of each frame;
• and
• an analysis module, implemented by the processor, configured to calculate popularity of the object features of the video segments according to the high level features and low level features and recommend at least one first recommended video segment to the user from the plurality of video segments according to the calculated popularity of the object features and the feedback information.

A video recommendation system, comprising:

• an interaction device configured to receive feedback information of videos from multiple users, wherein the videos comprise a plurality of video segments, and each video segment comprises a plurality of frames;
• a feature calculating module, implemented by a processor, configured to calculate at least one object feature of the plurality of video segments corresponding to the feedback information;
• and
• an analysis module, implemented by the processor, configured to calculate popularity of the at least one object feature of the plurality of video segments according to the high level features and the low level features, and recommend at least one recommended video segment from the plurality of video segments according to the popularity of the at least one object feature,
• wherein the at least one object feature comprises at least one high level feature and at least one low level feature of the plurality of video segments, and
• wherein the high level feature indicates object characteristics of the video segments or the frames close to explanation by human, and the low level features are close to the substantial properties of objects.