https://github.com/carsonfenimore/veye_mip327
Fork of some code from https://github.com/veyeimaging/raspberrypi.git
https://github.com/carsonfenimore/veye_mip327
Last synced: 3 months ago
JSON representation
Fork of some code from https://github.com/veyeimaging/raspberrypi.git
- Host: GitHub
- URL: https://github.com/carsonfenimore/veye_mip327
- Owner: carsonfenimore
- Created: 2020-12-17T13:29:04.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2020-12-17T13:59:03.000Z (over 5 years ago)
- Last Synced: 2025-02-26T03:44:37.252Z (over 1 year ago)
- Language: C
- Size: 19.5 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# veye_mip327
Fork of some code from https://github.com/veyeimaging/raspberrypi.git. The veyeimaging repo only provided H264 OR YUV streaming - but not both simultenaouesly.
This code allows simultaneous H264 and YUV frames capture using the IMX327 sensor as integrated by veye. On a raspberry pi 4 this enables full 1080p NAL and YUV capture. Even with YUV->rgb conversion, this process only consumes a small fraction (<10%) of total cpu - leaving the rest for vision / AI processing tasks.
Note that while it is possible to stream RGB, the internal bus of the raspberry pi is not able to push full 1080p @ 30fps. For this reason YUV is preferred for frame formats.
Note2: you can easily convert the yuv frames using opencv as follows:
```
cv::Mat yuv(frameHeight+frameHeight/2, frameWidth, CV_8UC1,(uchar *)yuvData);
```
## To use this code
```
git clone the https://github.com/veyeimaging/raspberrypi.git
git clone https://github.com/carsonfenimore/veye_mip327.git
cp veye_mip327/* raspberrypi/veye_raspcam/source
cd raspberrypi/veye_raspcam/source
./build.sh
./mipi327cap
```