Lintong
We also found that the test result on S3DIS using RandLA `pipeline.run_test()` has a low score 0.3 (supposed to be 0.7). When visualized, the segmentation result is not so great,...
They are Framos cameras IMX296 connected to Nvidia Jetson NX board with Framos [connector ](https://www.framos.com/en/products/fma-fc-150-60-v1a-22386). The cameras are configured as master and slaves through the driver, so they are triggered...
Thank you for the quick reply @Gofinge . We wanted to try OA-CNN since it's faster and lighter compared to PTv3, especially if we deploy on robots on a Nvidia...
Yes, I will do. Thank you for the good work and quick response!
@Gofinge, Hi, we have done some finetuning with our custom lidar data and it works fine. One challenge is converting PTv3 with TensorRT as we want to deploy on a...