search

NVIDIA-AI-IOT / Lidar_AI_Solution

A project demonstrating Lidar related AI solutions, including three GPU accelerated Lidar/camera DL networks (PointPillars, CenterPoint, BEVFusion) and the related libs (cuPCL, 3D SparseConvolution, YUV2RGB, cuOSD,).
Other
1.26k stars 220 forks source link

Could you specify which exact model of the NVIDIA Jetson Orin series was used in this project? #224

Open catalysts-higher opened 9 months ago

catalysts-higher commented 9 months ago

Could you specify which exact model of the NVIDIA Jetson Orin series was used to achieve the performance metrics detailed in the Readme? : Jetson AGX Orin 64GB Jetson AGX Orin Industrial Jetson AGX Orin 32GB Jetson Orin NX 16GB Jetson Orin NX 8GB Jetson Orin Nano 8GB Jetson Orin Nano 4GB

hopef commented 9 months ago

Jetson AGX Orin 64GB

guangqianzhang commented 8 months ago

there some question：

  1. if 8GB memory is qualified it， if not how many it need？
  2. Jetson Orin Nano 8GB sm=86，Whether it is qualified for the project？ whether must be the AGX orin？thank！
hopef commented 8 months ago

  1. if 8GB memory is qualified it， if not how many it need？ -> That would be enough to infer. But it would help if you considered removing some of the system footprint. If you turn CUDNN/cuBLAS off on TensorRT further, you can save even more memory.

  2. Jetson Orin Nano 8GB sm=86，Whether it is qualified for the project？ whether must be the AGX orin？thank！ -> Jetson Orin Nano 8GB sm=86 is ok for infer. AGX orin isn't necessary.