1. Projects

Learning3D: A Modern Library for Deep Learning on 3D Point Clouds Data.

2. Distance Metrics

2.1. Chamfer Distance



2.2. Earth Mover Distance



2.3. F-score



3. Sampling

3.1. Sampling from mesh



3.2. Sampling from point cloud

farthest point sampling



4. Normalization

def normalize_to_box(input):
    """
    normalize point cloud to unit bounding box
    center = (max - min)/2
    scale = max(abs(x))
    input: pc [N, P, dim] or [P, dim]
    output: pc, centroid, furthest_distance

    From https://github.com/yifita/pytorch_points
    """
    if len(input.shape) == 2:
        axis = 0
        P = input.shape[0]
        D = input.shape[1]
    elif len(input.shape) == 3:
        axis = 1
        P = input.shape[1]
        D = input.shape[2]
    else:
        raise ValueError()

    if isinstance(input, np.ndarray):
        maxP = np.amax(input, axis=axis, keepdims=True)
        minP = np.amin(input, axis=axis, keepdims=True)
        centroid = (maxP+minP)/2
        input = input - centroid
        furthest_distance = np.amax(np.abs(input), axis=(axis, -1), keepdims=True)
        input = input / furthest_distance
    elif isinstance(input, torch.Tensor):
        maxP = torch.max(input, dim=axis, keepdim=True)[0]
        minP = torch.min(input, dim=axis, keepdim=True)[0]
        centroid = (maxP+minP)/2
        input = input - centroid
        in_shape = list(input.shape[:axis])+[P*D]
        furthest_distance = torch.max(torch.abs(input).reshape(in_shape), dim=axis, keepdim=True)[0]
        furthest_distance = furthest_distance.unsqueeze(-1)
        input = input / furthest_distance
    else:
        raise ValueError()

    return input, centroid, furthest_distance

results matching ""

    No results matching ""