Skip to content

Tutorial - SAM (Segment Anything)

Let's run Meta's SAM on NVIDIA Jetson.

What you need

  1. One of the following Jetson:

    Jetson AGX Orin 64GB Jetson AGX Orin (32GB) Jetson Orin Nano Orin (8GB)⚠️1

  2. Running one of the following JetPack.5x

    JetPack 5.1.2 (L4T r35.4.1) JetPack 5.1.1 (L4T r35.3.1) JetPack 5.1 (L4T r35.2.1)

  3. Sufficient storage space (preferably with NVMe SSD).

    • 6.8GB for container image
    • Spaces for models

Set up a container for sam

Clone jetson-containers

See jetson-containers' sam package README for more infomation**

git clone https://github.com/dusty-nv/jetson-containers
cd jetson-containers
sudo apt update; sudo apt install -y python3-pip
pip3 install -r requirements.txt

How to start

Use run.sh and autotag script to automatically pull or build a compatible container image.

cd jetson-containers
./run.sh $(./autotag sam)

The container has a default run command (CMD) that will automatically start the Jupyter Lab server.

Open your browser and access http://<IP_ADDRESS>:8888.

The default password for Jupyter Lab is nvidia.

Run Jupyter notebook

In Jupyter Lab, navigate to notebooks and open automatic_mask_generator_example.py notebook.

Create a new cell at the top, insert the model download command below and run the cell.

!wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth

Then go through executing all the cells below Set-up.

Results

  1. The biggest vit_h (2.4GB) model may not ran due to OOM, but vit_l (1.1GB) runs on Jetson Orin Nano. ↩