Video3DStream

class quanser.multimedia.video.Video3DStream(stream)

A python wrapper for a video3d stream. You should not need to create an instance of this class yourself.

close()

Closes a stream.

Raises: MediaError – On non-zero return code. A suitable error message may be retrieved using get_error_message.

Example

Close an open depth stream.

>>> from quanser.multimedia import Video3D, ImageDataType, ImageFormat, Video3DStreamType
>>> video3d = Video3D("0")
>>> stream = video3d.stream_open(Video3DStreamType.DEPTH, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_GREYSCALE, ImageDataType.UINT8)
...
>>> # ...
...
>>> stream.close()
>>> video3d.close()

get_camera_intrinsics(intrinsics, coefficients)

Gets the camera intrinsics and the distortion model for the given stream.

Parameters

intrinsics (array_like) – A (3,3) array to fill with the camera intrinsic matrix.
coefficients (array_like) – A 5-element array to fill with the distortion model coefficients.

Raises

MediaError – On non-zero return code. A suitable error message may be retrieved using get_error_message.

Returns

An instance of a video3d distortion model.

Return type

Video3DDistortionModel

Example

Get the camera intrinsics and distortion model for the depth stream.

>>> from quanser.multimedia import Video3D, ImageDataType, ImageFormat, Video3DStreamType, Video3DDistortionModel
>>> video3d = Video3D("0")
>>> stream = video3d.stream_open(Video3DStreamType.DEPTH, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_GREYSCALE, ImageDataType.UINT8)
...
>>> intrinsics = np.zeros((3, 3), dtype=np.float32)
>>> coefficients = np.zeros((5), dtype=np.float32)
>>> model = stream.get_camera_intrinsics(intrinsics, coefficients)
...
>>> video3d.stop_streaming()
>>> stream.close()
>>> video3d.close()

get_depth_scale(data_type)

Gets the scale factor from depth expressed in the given data type to metres.

Parameters: data_type (ImageDataType) – The data type of the depth image for which to get the scale factor to metres.
Raises: MediaError – On non-zero return code. A suitable error message may be retrieved using get_error_message.

Example

Get the scale factor from a ImageDataType.UINT16 depth image to metres.

>>> from quanser.multimedia import Video3D, ImageDataType, ImageFormat, Video3DStreamType
>>> video3d = Video3D("0")
>>> depth_stream = video3d.stream_open(Video3DStreamType.DEPTH, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_GREYSCALE, ImageDataType.UINT16)
...
>>> depth_scale = depth_stream.get_depth_scale(ImageDataType.UINT16)
...
>>> video3d.stop_streaming()
>>> depth_stream.close()
>>> video3d.close()

get_extrinsics(to_stream, extrinsics)

Gets the extrinsics transformation from this stream to the given stream.

Parameters: extrinsics (array_like) – A (4,4) array to fill with the homogeneous transformation matrix representing the extrinsics.
Raises: MediaError – On non-zero return code. A suitable error message may be retrieved using get_error_message.

Example

Get the extrinsics transform from the current stream to the given stream.

>>> from quanser.multimedia import Video3D, ImageDataType, ImageFormat, Video3DStreamType
>>> video3d = Video3D("0")
>>> depth_stream = video3d.stream_open(Video3DStreamType.DEPTH, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_GREYSCALE, ImageDataType.UINT8)
>>> rgb_stream = video3d.stream_open(Video3DStreamType.COLOR, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_RGB, ImageDataType.UINT8)
...
>>> extrinsics = np.zeros((4, 4), dtype=np.float32)
>>> depth_stream.get_extrinsics(rgb_stream, extrinsics)
...
>>> video3d.stop_streaming()
>>> rgb_stream.close()
>>> depth_stream.close()
>>> video3d.close()

get_frame()

Get the latest frame for the given stream. If no frame is available, then it raises QERR_WOULD_BLOCK. Only one frame may be retrieved at a time and the frame must be released before getting a new frame.

Raises: MediaError – On non-zero return code. A suitable error message may be retrieved using get_error_message.
Returns: An instance of a video3d frame or None, if no frame was available.
Return type: Video3DFrame

Example

Capture and process 1000 frames from the depth stream.

>>> from quanser.multimedia import Video3D, ImageDataType, ImageFormat, Video3DStreamType
>>> video3d = Video3D("0")
>>> stream = video3d.stream_open(Video3DStreamType.DEPTH, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_GREYSCALE, ImageDataType.UINT8)
...
>>> video3d.start_streaming()
>>> for i in range(1000):
...     frame = stream.get_frame()
...     # Process...
...     frame.release()
...
>>> video3d.stop_streaming()
>>> stream.close()
>>> video3d.close()

set_properties(properties, num_properties, values)

Sets the value of one or more properties of a stream. For boolean options, use a value of 0.0 or 1.0. All other options should have a range of 0.0 to 1.0 inclusive. The function will do the appropriate scaling for the internal camera settings. Any values outside the range will be saturated to lie within the range.

Parameters

properties (array_like) – An array of Video3DProperty to set.
num_properties (int) – The number of properties in the properties array.
values (array_like) – An array of values. Each element in the values array corresponds to the same element in the properties array.

Raises

MediaError – On non-zero return code. A suitable error message may be retrieved using get_error_message.

Example

Set the brightness to 100% and the contrast to 50% on a depth stream.

>>> import numpy as np
>>> from quanser.multimedia import Video3D, ImageDataType, ImageFormat, Video3DProperty, Video3DStreamType
>>> video3d = Video3D("0")
>>> stream = video3d.stream_open(Video3DStreamType.DEPTH, 0, 30.0, 1280, 720,
...                              ImageFormat.ROW_MAJOR_GREYSCALE, ImageDataType.UINT8)
...
>>> properties = np.array([Video3DProperty.BRIGHTNESS, Video3DProperty.CONTRAST], dtype=np.int32)
>>> num_properties = len(properties)
>>> values = np.array([1.0, 0.5], dtype=np.float64)
>>> stream.set_properties(properties, num_properties, values)
>>> # ...
...
>>> stream.close()
>>> video3d.close()