tf.image.extract_jpeg_shape

tf.image.extract_jpeg_shape(
    contents,
    output_type=tf.int32,
    name=None
)

Defined in tensorflow/python/ops/gen_image_ops.py.

Extract the shape information of a JPEG-encoded image.

This op only parses the image header, so it is much faster than DecodeJpeg.

Args:

• contents: A Tensor of type string. 0-D. The JPEG-encoded image.
• output_type: An optional tf.DType from: tf.int32, tf.int64. Defaults to tf.int32. (Optional) The output type of the operation (int32 or int64). Defaults to int32.
• name: A name for the operation (optional).

Returns:

A Tensor of type output_type.