asmagic 0.1.0

AR Sensor Data Magic Library - Subscribe to AR sensor data streams

asmagic

A lightweight Python library for subscribing to AR sensor data streams via ZeroMQ.

Installation

pip install asmagic

For image display support:

pip install asmagic[image]

Quick Start

from asmagic import ARDataSubscriber

# Create subscriber
sub = ARDataSubscriber("192.168.1.100")

# Get data frame
data = sub.get()
if data:
    print(data.timestamp)
    print(data.velocity)
    print(data.local_pose)
    print(data.camera_intrinsics)

sub.close()

Usage

Continuous Data Streaming

from asmagic import ARDataSubscriber

sub = ARDataSubscriber("192.168.1.100")

try:
    while True:
        data = sub.get()
        if data:
            print(f"Velocity: {data.velocity}")
            print(f"Local pose: {data.local_pose}")
            print(f"Timestamp: {data.timestamp}")
except KeyboardInterrupt:
    pass
finally:
    sub.close()

Image Display

Requires opencv-python and Pillow:

pip install asmagic[image]

from asmagic import ARDataSubscriber
import cv2

sub = ARDataSubscriber("192.168.1.100")

try:
    while True:
        data = sub.get()
        if data:
            # Display both color and depth images
            data.show_images()

            # Or display individually
            # data.show_color()
            # data.show_depth()

        if cv2.waitKey(1) == 27:  # ESC to exit
            break
finally:
    sub.close()
    cv2.destroyAllWindows()

Accessing Individual Fields

from asmagic import ARDataSubscriber

sub = ARDataSubscriber("192.168.1.100")

# Get specific data
velocity = sub.get_velocity()
pose = sub.get_local_pose()
timestamp = sub.get_timestamp()

sub.close()

API Reference

ARDataSubscriber

Constructor:

ARDataSubscriber(ip, port=8000, hwm=1, conflate=True, verbose=False)

Parameters:

• ip (str): IP address of the AR device
• port (int): Port number (default: 8000)
• hwm (int): High water mark for ZMQ socket (default: 1)
• conflate (bool): Enable message conflation (default: True)
• verbose (bool): Print connection info (default: False)

Methods:

Method	Returns	Description
get(timeout=1000)	ARFrame or None	Get latest data frame
get_timestamp(timeout=1000)	float or None	Get timestamp
get_velocity(timeout=1000)	np.ndarray or None	Get velocity
get_local_pose(timeout=1000)	np.ndarray or None	Get local pose
get_global_pose(timeout=1000)	np.ndarray or None	Get global pose
get_camera_intrinsics(timeout=1000)	np.ndarray or None	Get camera intrinsics
get_color_image(timeout=1000)	bytes or None	Get color image bytes
get_depth_image(timeout=1000)	np.ndarray or None	Get depth image array
close()	None	Close connection

ARFrame

Data object returned by get():

Properties:

Property	Type	Description
timestamp	float	Frame timestamp
velocity	np.ndarray	Velocity vector
local_pose	np.ndarray	Local pose matrix
global_pose	np.ndarray	Global pose matrix
camera_intrinsics	np.ndarray	Camera intrinsic parameters
color_img	bytes	Color image bytes (JPEG/PNG)
depth_img	bytes	Raw depth image bytes
depth_width	int	Depth image width
depth_height	int	Depth image height
has_color_image	bool	Whether color image is available
has_depth_image	bool	Whether depth image is available

Methods:

Method	Returns	Description
get_depth_array()	np.ndarray	Get depth as numpy array (uint16)
show_color(window_name)	bool	Display color image
show_depth(window_name, colormap)	bool	Display depth image
show_images(show_color, show_depth)	tuple	Display both images

Requirements

• Python >= 3.8
• numpy >= 1.20.0
• pyzmq >= 22.0.0
• protobuf >= 4.0.0

Optional (for image display):

• opencv-python >= 4.5.0
• Pillow >= 8.0.0

License

MIT License