What is R and T?

[UestcJay]

when I use PointsRenderer, :
cameras = FoVOrthographicCameras(device=device, R=R, T=T, znear=0.01), is the R and T extrinsics of the camera? I want to render point cloud in world to a view, Is this all right?
` R = torch.from_numpy(extrinsics_ref[:3,:3][np.newaxis,:,:])

    T = torch.from_numpy(extrinsics_ref[:3,3][np.newaxis,:])

    cameras = FoVPerspectiveCameras(device='cpu', R=R, T=T, znear=0.01)
    raster_settings = PointsRasterizationSettings(
        image_size=(128,160),
        radius=0.03,
        points_per_pixel=10
    )

    rasterizer = PointsRasterizer(cameras=cameras, raster_settings=raster_settings)

    renderer = PointsRenderer(
        rasterizer=rasterizer,
        compositor=AlphaCompositor()
    )

    verts = torch.from_numpy(xyz_ref.T.astype(np.float32)).to('cpu')

    rgb = torch.from_numpy(resized_img_src).to('cpu')

    point_cloud = Pointclouds(points=[verts], features=[rgb])

    images = renderer(point_cloud)` 

[bottler]

Yes, R and T are the extrinsics, defining the camera location. https://pytorch3d.org/docs/cameras may be helpful.

[UestcJay]

这是来自QQ邮箱的自动回复邮件。你好，你的来件我已收到，谢谢。