GStreamer Based Image Signal Processor - Performance - imx8

CSC Benchmarking

The following section describes some relevant performance measurements for the RGBA to NV12 color space conversion for a few standard resolutions. It begins with a summary of the results detailed on the following sections, as well as the pipelines used to capture such results.

Pipeline

All the measurements below were made using the following minimal pipeline.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 videotestsrc is-live=true ! "video/x-raw,format=(string)RGBA,width=$WIDTH,height=$HEIGHT,framerate=(fraction)30/1" ! queue ! ispclcsc ! fakesink -v

Summary

In this summary you can find measurements related with CPU usage, memory, execution time and maximum framerate supported by the CSC algorithm.

Table 1. Performance for several standard resolutions at 30fps

Resolution	CPU usage (%)	Memory Bandwidth(GHz)	Execution Time (ms)
640x480@30fps	4	18664	2.24
1280x720@30fps	10	19484	6.19
1920x1080@30fps	14	21424	13.17

Table 2. Maximum framerate at standard resolutions

Resolution	Maximum framerate (fps) *
640x480	333
1280x720	120
1920x1080	55

*Note: take into consideration that these values were taken using the videotestsrc element, and may decrease with a camera source capture.

CPU usage

These measurements were taken with top app to monitor the behavior of the pipeline while running.

CPU usage for CSC kernel.

Memory usage

These measurements were taken with meminfo file to see the behavior of the pipeline while running.

Memory consumption by CSC kernel.

Execution time

These measurements were taken using some timestamps before and after the processing to measure just the execution time for the CSC kernel algorithm on each frame and hence avoiding possible false measurements from the GStreamer plugin. The image below shows the CSC execution time by each frame for each resolution.

Execution time for different resolutions on the CSC kernel.

Maximum Framerate

These measurements were taken with the gst-perf tool to monitor the behavior of the pipeline while running. When using videotestsrc we are able to process as many buffers as the element can handle, hence the maximum value reported. The image below shows the maximum framerate that the CSC element supports, which depends on the execution time of the CSC kernel algorithm.

The following pipeline was used to test the maximum framerate, please note the use of videotestsrc.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 videotestsrc ! "video/x-raw,format=(string)RGBA,width=$WIDTH,height=$HEIGHT,framerate=(fraction)30/1" ! queue ! ispclcsc ! fakesink -v

Framerate for different resolutions on the CSC kernel.

Debayer Benckmarking

The following section describes some relevant performance measurements during debayering for a few standard resolutions. It begins with a summary of the results detailed on the following sections, as well as the pipelines used to capture such results.

Pipeline

All the measurements below were made using the following minimal pipeline.

WIDTH=1920
HEIGHT=1080

$ gst-launch-1.0 v4l2src device=/dev/video0 ! video/x-bayer,format=bggr10,width=$WIDTH,height=$HEIGHT,framerate=30/1 ! ispclbayer10to8 ! queue ! ispcldebayer ! fakesink -v

Summary

In this summary you can find measurements related with CPU usage, memory, execution time and maximum framerate supported by the debayer algorithm.

Table 1. Performance for several standard resolutions at 30fps

Resolution	CPU usage (%)	Memory Bandwidth(GHz)	Execution Time (ms)
640x480@30fps	2	20084	3.54
1280x720@30fps	2	22564	9.89
1920x1080@30fps	3	23212	21.46

Table 2. Maximum framerate at standard resolutions

Resolution	Maximum framerate (fps)*
640x480	266
1280x720	101
1920x1080	46

*Note: take into consideration that these values were taken using the videotestsrc element, and may decrease with a camera source capture.

CPU usage

These measurements were taken with top app to see the behavior of the pipeline while running at 30fps.

Cpu usage debayering for different resolutions at 30fps.

Memory usage

These measurements were taken with meminfo file to see the behavior of the pipeline while running at 30fps.

Memory usage debayering for different resolutions at 30fps.

Execution time

These measurements were taken putting some timestamps before and after the processing to measure just the execution time for the debayer kernel algorithm by each frame and avoiding possible false measurements from the GStreamer plugin. The image below shows the debayer execution time by each frame for each resolution.

Execution time debayering for different resolutions at 30fps.

Maximum Framerate

These measurements were taken with gstperf tool to see the behavior of the pipeline while running, with videotestsrc we are able to push buffers as much as possible to the element. The image below shows the maximum framerate that the debayer element supports. It depends on the execution time of the debayer kernel algorithm.

The following pipeline was used to test the maximum framerate, please note the use of videotestsrc.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 videotestsrc ! "video/x-bayer,format=bggr,width=$WIDTH,height=$HEIGHT,framerate=30/1" ! queue ! ispcldebayer ! perf ! fakesink -v

Maximum framerate debayering for different resolutions.

AWB Benchmarking

The following section describes some relevant performance measurements for the auto-white balance for a few standard resolutions. It begins with a summary of the results detailed on the following sections, as well as the pipelines used to capture such results.

Pipeline

All the measurements below were made using the following minimal pipeline.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 v4l2src device=/dev/video0 ! video/x-bayer,format=bggr10,width=$WIDTH,height=$HEIGHT,framerate=30/1 ! ispclbayer10to8 ! queue ! ispclawb ! fakesink -v

Summary

In this summary you can find measurements related with CPU usage, memory, execution time and maximum framerate supported by the auto-white balance algorithm.

Table 1. Performance for several standard resolutions at 30fps

Resolution	CPU usage (%)	Memory Bandwidth(GHz)	Execution Time (ms)
640x480@30fps	10		3.11
1280x720@30fps	12		4.54
1920x1080@30fps	16		8.52

Table 2. Maximum framerate at standard resolutions

Resolution	Maximum framerate (fps) *
640x480	365
1280x720	135
1920x1080	62

*Note: take into consideration that these values were taken using the videotestsrc element, and may decrease with a camera source capture.

CPU usage

These measurements were taken with top app to see the behavior of the pipeline while running at 30fps.

Memory usage

These measurements were taken with meminfo file to see the behavior of the pipeline while running at 30fps.

Execution time

These measurements were taken putting some timestamps before and after the processing to measure just the execution time for the debayer kernel algorithm by each frame and avoiding possible false measurements from the GStreamer plugin. The image below shows the autowhite balance execution time by each frame for each resolution.

Execution time for Autowhite balance for different resolutions.

Maximum Framerate

These measurements were taken with gstperf tool to see the behavior of the pipeline while running, with videotestsrc we are able to push buffers as much as possible to the element. The image below shows the maximum framerate that the debayer element supports. It depends on the execution time of the debayer kernel algorithm.

The following pipeline was used to test the maximum framerate, please note the use of videotestsrc.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 videotestsrc ! "video/x-bayer,format=bggr,width=$WIDTH,height=$HEIGHT,framerate=30/1" ! queue ! ispclawb ! perf ! fakesink -v

Maximum framerate for Autowhite balance for different resolutions.

Histogram Equalization Benchmarking

The following section describes some relevant performance measurements for the Histogram Equalization for a few standard resolutions. It begins with a summary of the results detailed on the following sections, as well as the pipelines used to capture such results.

Pipeline

All the measurements below were made using the following minimal pipeline.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 v4l2src device=/dev/video0 ! video/x-bayer,format=bggr10,width=$WIDTH,height=$HEIGHT,framerate=30/1 ! ispclbayer10to8 ! queue ! ispcleq ! fakesink -v

Summary

In this summary you can find measurements related with CPU usage, memory, execution time and maximum framerate supported by the Histogram Equalization algorithm.

Table 1. Performance for several standard resolutions at 30fps

Resolution	CPU usage (%)	Memory Bandwidth(GHz)	Execution Time (ms)
640x480@30fps	10		4.28
1280x720@30fps	12		6.07
1920x1080@30fps	16		10.95

Table 2. Maximum framerate at standard resolutions

Resolution	Maximum framerate (fps)*
640x480	345
1280x720	135
1920x1080	62

*Note: take into consideration that these values were taken using the videotestsrc element, and may decrease with a camera source capture.

CPU usage

These measurements were taken with top app to see the behavior of the pipeline while running at 30fps.

Memory usage

These measurements were taken with meminfo file to see the behavior of the pipeline while running at 30fps.

Execution time

These measurements were taken putting some timestamps before and after the processing to measure just the execution time for the histogram equalization kernel algorithm by each frame and avoiding possible false measurements from the GStreamer plugin. The image below shows the histogram equalization execution time by each frame for each resolution.

Execution time of Histogram Equalization for different resolutions.

Maximum Framerate

These measurements were taken with gstperf tool to see the behavior of the pipeline while running, with videotestsrc we are able to push buffers as much as possible to the element. The image below shows the maximum framerate that the debayer element supports. It depends on the execution time of the debayer kernel algorithm.

The following pipeline was used to test the maximum framerate, please note the use of videotestsrc.

$ WIDTH=1920
$ HEIGHT=1080

$ gst-launch-1.0 videotestsrc ! "video/x-bayer,format=bggr,width=$WIDTH,height=$HEIGHT,framerate=30/1" ! queue ! ispcleq ! perf ! fakesink -v

Maximum framerate of Histogram Equalization for different resolutions.