Function _get_param_spatial_crop

slowfast/datasets/transform.py:535–574

Given scale, ratio, height and width, return sampled coordinates of the videos.

(
    scale, ratio, height, width, num_repeat=10, log_scale=True, switch_hw=False
)

def _get_param_spatial_crop(
    scale, ratio, height, width, num_repeat=10, log_scale=True, switch_hw=False
):
    """
    Given scale, ratio, height and width, return sampled coordinates of the videos.
    """
    for _ in range(num_repeat):
        area = height * width
        target_area = random.uniform(*scale) * area
        if log_scale:
            log_ratio = (math.log(ratio[0]), math.log(ratio[1]))
            aspect_ratio = math.exp(random.uniform(*log_ratio))
        else:
            aspect_ratio = random.uniform(*ratio)

        w = int(round(math.sqrt(target_area * aspect_ratio)))
        h = int(round(math.sqrt(target_area / aspect_ratio)))

        if np.random.uniform() < 0.5 and switch_hw:
            w, h = h, w

        if 0 < w <= width and 0 < h <= height:
            i = random.randint(0, height - h)
            j = random.randint(0, width - w)
            return i, j, h, w

    # Fallback to central crop
    in_ratio = float(width) / float(height)
    if in_ratio < min(ratio):
        w = width
        h = int(round(w / min(ratio)))
    elif in_ratio > max(ratio):
        h = height
        w = int(round(h * max(ratio)))
    else:  # whole image
        w = width
        h = height
    i = (height - h) // 2
    j = (width - w) // 2
    return i, j, h, w


def random_resized_crop(