search

mzahana / jetson_vins_fusion_docker

Docker files and scripts to setup and run VINS-FUSION-gpu on NVIDIA jetson boards inside a docker container.
BSD 3-Clause "New" or "Revised" License
32 stars 4 forks source link

Docker image on SD card (Jetson TX2) #2

Open Pallav1299 opened 2 years ago

Pallav1299 commented 2 years ago

Thanks for the docker image @mzahana.

I followed the procedure mentioned in the README but due to large image size and storage constraints on Jetson TX2, I added the built docker image on external SD card. While testing Vins_fusion_gpu I faced latency issues. I suspect that, having the docker image on the SD card is leading to this latency. What are your views on the same?

What is the size of built docker image for you? Have you used an external storage for your setup?

mzahana commented 2 years ago

@Pallav1299 unfortunately I don't have TX2 to test with. This image was successfully tested on Xavier NX with SSD drive. The image size is large ~14GB, but I think it can be reduced significantly if the source file of OpenCV is removed after installation inside the image.

Pallav1299 commented 2 years ago

@mzahana, unfortunately I could't make enough space for the docker image on device's flash memory. I am still sceptical if having the VINS-GPU docker image on SD card can have any viable impact on VINS performance. What are your views on the same?

mzahana commented 2 years ago

@Pallav1299 I would suggest to get good quality high speed SD card and test.

Pallav1299 commented 2 years ago

@mzahana, I've tested with SSD as well. Doesn't seem to be an issue with having docker image on SD card. What sensor rates did you use for Camera and IMU data? Also, what values did you use for max_cnt(max number of features to track) and min_dist(minimum distance between two features) in VINS?