SYSTEMS AND METHODS FOR SUBJECT-ORIENTED COMPRESSION
Examples of the present disclosure relate to performing subject oriented compression. A content file, such as a video file, may be received. One or more subjects of interest may be identified in the content file. The identified subjects of interest may be associated with a quantization value that is less than a quantization value associated with the rest of the content. When the content is compressed/encoded, the subjects of interest are compressed/encoded using their associated quantization value while the rest of the content is compressed/encoded using a larger quantization value.
1 . A method of performing subject-oriented compression, the method comprising:
identifying a subject of interest in an image;
compressing the subject of interest using a first quantization value; and
compressing the remainder of the image using a second quantization value, wherein the second quantization value is greater than the first quantization value.
2 . The method of claim 1 , wherein identifying the subject of interest comprises automatically identifying the subject of interest based on at least one characteristic of the subject of interest.
3 . The method of claim 1 , wherein identifying the subject of interest further comprises:
displaying a frame in a graphical user interface (GUI); and
receiving an indication of the subject of interest via the GUI.
4 . The method of claim 3 , wherein receiving the indication of the subject of interest comprises receiving a click-and-drag input.
5 . The method of claim 1 , wherein the subject of interest is identified by a bounding box.
6 . The method of claim 1 , further comprising:
identifying a second subject of interest; and
compressing the second subject of interest using the first quantization value.
7 . The method of claim 6 , wherein the first subject of interest and the second subject of interest overlap.
8 . The method of claim 6 , further comprising:
identifying a third subject of interest; and
compressing the third subject of interest using a third quantization value, wherein the third quantization value is different from the first quantization value and the second quantization value.
9 . A system comprising:
at least one processor; and
memory encoding computer executable instructions that, when executed by the at least one processor, perform a method comprising:
receiving a video;
creating at least one metadata file;
identifying at least one subject of interest;
associating a quantization value with the at least one subject of interest;
tracking the at least one subject of interest; and
saving metadata generated during tracking to the at least one metadata file.
10 . The system of claim 9 , wherein creating at least one metadata file comprises:
creating a first metadata file, wherein the first metadata file comprises data about the at least one subject of interest; and
creating a second metadata file, wherein the second metadata file comprises data about at least one quantization value;
11 . The system of claim 10 , wherein saving metadata generated during the tracking comprises saving metadata about the at least one subject of interest to the first metadata file.
12 . The system of claim 11 , wherein the metadata comprises saved to the first metadata file comprises:
data identifying a frame;
data identifying a location for the at least one subject of interest; and
a segment identifier.
13 . The system of claim 9 , wherein tracking the at least one subject of interest comprises performing feature tracking.
14 . The system of claim 13 , wherein the method further comprises:
determining whether the at least one subject of interest is lost; and
generating a notification that the at least one subject of interest is not available for a specific frame.
15 . The system of claim 14 , wherein the method further comprises receiving input identifying the at least one subject of interest in the specific frame.
16 . The system of claim 9 , wherein the method further comprises associating at least one quantization value with the at least one subject of interest, wherein the at least one quantization value is less than a background quantization value.
17 . The system of claim 16 , wherein the method further comprises compressing the video, wherein the at least one subject of interest is compressed using the at least one quantization value and the rest of the video is compressed using the background quantization value.
18 . A computer storage medium encoding computer executable instructions that, when executed by at least one processor, perform a method comprising:
receiving a video;
creating at least one metadata file;
identifying at least one subject of interest;
associating a quantization value with the at least one subject of interest;
tracking the at least one subject of interest;
saving metadata generated during tracking to the at least one metadata file; and
performing subject oriented compression on the video using the at least one metadata file.
19 . The computer storage medium of claim 18 , wherein the method further comprises associating at least one quantization value with the at least one subject of interest, wherein the at least one quantization value is less than a background quantization value.
20 . The computer storage medium of claim 19 , wherein performing subject oriented compression further comprises compressing the at least one subject of interest using the at least one quantization value.