Abstract: Embodiments provide a system and method for filtering speech in a video. Speech in video may contain objectionable or profane words that need to be filtered. To ascertain whether a word or phrase is objectionable, the contextual information from surrounding words and the contextual information from detected objects and scenes in the video are used. Unwanted words may be filtered or collected and presented to the user.

Abstract: Techniques for detecting epileptic triggers in video content by storing a first sequence of frames of a video stream in a first buffer. The technique can further comprise generating a second sequence of subframes and storing the second sequence of subframes in a second buffer. The technique can further comprise generating average intensity values for the second sequence of subframes in the second buffer and calculating intensity changes between consecutive subframes in the second sequence of subframes based on the average intensity values. The technique can further comprise determining that a number of intensity changes between consecutive subframes in the second sequence of subframes exceeds an oscillation threshold and tagging the frames in the first buffer corresponding to the subframes in the second buffer as hazardous.

Abstract: Embodiments provide a system and method for filtering speech in a video. Speech in video may contain objectionable or profane words that need to be filtered. To ascertain whether a word or phrase is objectionable, the contextual information from surrounding words and the contextual information from detected objects and scenes in the video are used. Unwanted words may be filtered or collected and presented to the user.

Abstract: Embodiments provide a system and method for filtering speech in a video. Speech in video may contain objectionable words that need to be filtered. To ascertain whether a word or phrase is objectionable, the contextual information from surrounding words and the contextual information from detected objects and scenes in the video are used. Unwanted words may be filtered or collected and presented to the user.

Abstract: Techniques for detecting epileptic triggers in video content by storing a first sequence of frames of a video stream in a first buffer. The technique can further comprise generating a second sequence of subframes and storing the second sequence of subframes in a second buffer. The technique can further comprise generating average intensity values for the second sequence of subframes in the second buffer and calculating intensity changes between consecutive subframes in the second sequence of subframes based on the average intensity values. The technique can further comprise determining that a number of intensity changes between consecutive subframes in the second sequence of subframes exceeds an oscillation threshold and tagging the frames in the first buffer corresponding to the subframes in the second buffer as hazardous.

Abstract: Techniques for detecting epileptic triggers in video content by storing a first sequence of frames of a video stream in a first buffer. The technique can further comprise generating a second sequence of subframes and storing the second sequence of subframes in a second buffer. The technique can further comprise generating average intensity values for the second sequence of subframes in the second buffer and calculating intensity changes between consecutive subframes in the second sequence of subframes based on the average intensity values. The technique can further comprise determining that a number of intensity changes between consecutive subframes in the second sequence of subframes exceeds an oscillation threshold and tagging the frames in the first buffer corresponding to the subframes in the second buffer as hazardous.

Abstract: Techniques for detecting epileptic triggers in video content by storing a first sequence of frames of a video stream in a first buffer. The technique can further comprise generating a second sequence of subframes and storing the second sequence of subframes in a second buffer. The technique can further comprise generating average intensity values for the second sequence of subframes in the second buffer and calculating intensity changes between consecutive subframes in the second sequence of subframes based on the average intensity values. The technique can further comprise determining that a number of intensity changes between consecutive subframes in the second sequence of subframes exceeds an oscillation threshold and tagging the frames in the first buffer corresponding to the subframes in the second buffer as hazardous.

Abstract: Embodiments provide a system and method for filtering speech in a video. Speech in video may contain objectionable words that need to be filtered. To ascertain whether a word or phrase is objectionable, the contextual information from surrounding words and the contextual information from detected objects and scenes in the video are used. Unwanted words may be filtered or collected and presented to the user.