Preview streaming of video data

Patent metadata

Field	Value
Publication number	US-9635307-B1
Application number	US-201514974681-A
Country	US
Kind code	B1
Filing date	Dec 18, 2015
Priority date	Dec 18, 2015
Publication date	Apr 25, 2017
Grant date	Apr 25, 2017

Abstract

A system and method for generating preview data from video data and using the preview data to select portions of the video data or determine an order with which to upload the video data. The system may sample video data to generate sampled video data and may identify portions of the sampled video data having complexity metrics exceeding a threshold. The system may upload a first portion of the video data corresponding to the identified portions while omitting a second portion of the video data. The system may determine an order with which to upload portions of the video data based on a complexity of the video data. Therefore, portions of the video data that may require additional processing after being uploaded may be prioritized and uploaded first. As a result, a latency between the video data being uploaded and a video summarization being received is reduced.

First claim

What is claimed is:

1. A computer-implemented method for uploading video data, the method comprising:
capturing first video data having a first frame rate of 30 frames per second;
determining the first video data comprises a first portion associated with a first time;
determining the first video data comprises a second portion associated with a second time captured after the first time;
sampling the first video data to generate sampled video data having a second frame rate of 1 frame per second;
determining a plurality of color histograms, including determining a first color histogram that corresponds to a first video frame in the sampled video data;
determining a plurality of similarity scores, including:
determining a first similarity score, which corresponds to a first similarity between the first video frame and a second video frame in the sampled video data, by comparing the first color histogram to a second color histogram corresponding to the second video frame; and
determining a second similarity score, which corresponds to a second similarity between the first video frame and a third video frame in the sampled video data, by comparing the first color histogram to a third color histogram corresponding to the third video frame;

determining that a first group of similarity scores of the plurality of similarity scores are within a first range;
determining a first series of contiguous video frames corresponding to the first group of similarity scores;
determining that a second group of similarity scores of the plurality of similarity scores are within a second range;
determining a second series of contiguous video frames corresponding to the second group of similarity scores, the second series beginning with a fourth video frame in the sampled video data;
determining a plurality of transition points in the sampled video data, including determining a first transition point that corresponds to an intersection between the first series and the second series by determining that a third similarity score, which corresponds to a third similarity between the first video frame and the fourth video frame, is below a threshold;
determining a first number of the plurality of transition points corresponding to the first portion;
determining a second number of the plurality of transition points corresponding to the second portion;
determining that the second number of transition points exceeds the first number of transition points; and
sending, to a remote device, the second portion prior to sending the first portion.

2. The computer-implemented method of claim 1 , further comprising:
determining, using a first period of time, that the first video data comprises a plurality of portions including the first portion and the second portion;
determining the first portion associated with the first time, wherein a length of the first portion corresponds to the first period of time;
determining a third portion of the plurality of portions, wherein a length of the third portion corresponds to the first period of time;
determining a third number of the plurality of transition points corresponding to the third portion;
determining that the third number of transition points is below a threshold; and
determining not to send the third portion to the remote device.

3. The computer-implemented method of claim 1 , further comprising:
determining that a third group of similarity scores of the plurality of similarity scores are above a threshold;
determining a third series of contiguous video frames corresponding to the third group of similarity scores;
selecting the third series as the first portion;
determining that a fourth group of similarity scores of the plurality of similarity scores are above the threshold;
determining a fourth series of contiguous video frames corresponding to the fourth group of similarity scores, wherein the fourth series is longer than the third series; and
selecting the fourth series as the second portion.

4. The computer-implemented method of claim 1 , further comprising:
determining, using the plurality of transition points, a plurality of sections of the first video data, including determining a first section of the plurality of sections, which corresponds to the first transition point and is associated with a first period of time;
sending the plurality of sections to the remote device;
receiving, from the remote device, a request for a third portion of the first video data including the first section; and
sending the third portion to the remote device.

5. A computer-implemented method, comprising:
receiving first video data having a first frame rate;
determining the first video data comprises a first portion associated with a first time;
determining the first video data comprises a second portion associated with a second time that indicates the second portion being captured after the first portion;
generating sampled video data having a second frame rate lower than the first frame rate;
generating preview data comprising the sampled video data and at least one of:
annotation data associated with the first video data, audio data corresponding to the first video data or characteristic data indicating characteristics of the audio data; and
determining, based at least in part on the preview data, to send the second portion to a remote device prior to sending the first portion.

6. The computer-implemented method of claim 5 , further comprising:
determining a first similarity score that corresponds to a first similarity between a first video frame and a second video frame in the sampled video data;
determining a second similarity score that corresponds to a second similarity between the first video frame and a third video frame in the sampled video data;
determining, using the first similarity score and the second similarity score, a first series of contiguous video frames;
determining a second series of contiguous video frames;
determining a first transition point that corresponds to a transition between the first series and the second series;
determining a first number of transition points that correspond to the first portion;
determining a second number of transition points that correspond to the second portion;
determining that the second number exceeds the first number; and
determining to send the second portion to the remote device prior to sending the first portion.

7. The computer-implemented method of claim 5 , further comprising:
determining, using a first period of time, that the first video data comprises a plurality of portions including the first portion and the second portion;
determining the first portion associated with the first time, wherein a length of the first portion corresponds to the first period of time;
determining a third portion of the plurality of portions, wherein a length of the third portion corresponds to the first period of time;
determining a third number of the transition points corresponding to the third portion;
determining that the third number is below a threshold; and
determining not to send the third portion to the remote device.

8. The computer-implemented method of claim 5 , further comprising:
determining a first similarity score that corresponds to a first similarity between a first video frame and a second video frame in the sampled video data;
determining a second similarity score that corresponds to a second similarity between the first video frame and a third video frame in the sampled video data;
determining a first group of similarity scores that are above a threshold;
determining a first series of contiguous video frames corresponding to the first group of similarity scores;
selecting the first series as the first p

Assignees

• Amazon Tech Inc

Inventors

• Mysore Vijaya Kumar Rohith
• Tyagi Ambrish
• Rao Yadunandana Nagaraja
• Lakhani Suresh Bholabhai
• Agrawal Amit Kumar

Classifications

• G06V20/70

  Labelling scene content, e.g. deriving syntactic or semantic representations · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title

• H04N21/2743

  Video hosting of uploaded data from client · CPC title

• H04N21/4223

  Cameras (H04N23/00 takes precedence) · CPC title

• H04N7/0127

  by changing the field or frame frequency of the incoming video signal, e.g. frame rate converter · CPC title