5Input sources

Faery supports multiple input sources, each creating different types of streams. Understanding which input creates which stream type is crucial for building effective processing pipelines.

5.1Stream type summary¶

5.2Input source details¶

5.2.1📁 Files¶

Stream Type: FiniteEventsStream (Finite + Irregular)

Supported Formats: AEDAT4, ES, Prophesee RAW, DAT, EVT, CSV

Characteristics:

• Known duration and size

• Can be converted to arrays with .to_array()

• Perfect for video generation after regularization

• Reproducible processing

Example:

# File → Finite → Regular → Video
faery.events_stream_from_file("data.aedat4") \
    .regularize(frequency_hz=30.0) \
    .render(decay="exponential", tau="00:00:00.200000", colormap=faery.colormaps.devon) \
    .to_file("output.mp4")

5.2.2📷 Event Cameras¶

Stream Type: EventsStream (Infinite + Irregular)

Supported Hardware: DVXplorer, DAVIS346, EVK4, EVK3 HD

Characteristics:

• Continuous data stream

• Requires .time_slice() or .event_slice() for video output

• Real-time processing capabilities

• Live monitoring and streaming

Example:

# Camera → Slice → Finite → Regular → Video
faery.events_stream_from_camera() \
    .time_slice(0 * faery.s, 10 * faery.s) \
    .regularize(frequency_hz=30.0) \
    .render(decay="exponential", tau="00:00:00.200000", colormap=faery.colormaps.starry_night) \
    .to_file("camera_recording.mp4")

5.2.3🌐 UDP Streams¶

Stream Type: EventsStream (Infinite + Irregular)

Supported Formats: t64_x16_y16_on8, t32_x16_y15_on1

Characteristics:

• Network-based event streaming

• Low-latency for distributed systems

• Requires dimensions specification

• Can be finite or infinite depending on sender

Example:

# UDP → Process → Stream to another UDP endpoint
faery.events_stream_from_udp(
    dimensions=(640, 480),
    address=("localhost", 7777)
).remove_off_events() \
 .to_udp(("processing-server", 8888))

5.2.4🔢 Arrays (Python)¶

Stream Type: FiniteEventsStream (Finite + Irregular)

Data Source: NumPy arrays with faery.EVENTS_DTYPE

Characteristics:

• Programmatically generated data

• Perfect for testing and validation

• Known size and duration

• Synthetic data generation

Example:

# Array → Direct video generation
events = create_synthetic_events()  # Your function
faery.events_stream_from_array(events, dimensions=(640, 480)) \
    .regularize(frequency_hz=24.0) \
    .render(decay="exponential", tau="00:00:00.100000", colormap=faery.colormaps.batlow) \
    .to_file("synthetic.mp4")

5.2.5📥 Standard Input¶

Stream Type: FiniteEventsStream (Finite + Irregular)

Supported Format: CSV only

Characteristics:

• Command-line pipeline integration

• Finite by nature (piped data has an end)

• Configurable CSV parsing

• Shell scripting integration

Example:

# Shell pipeline → Faery processing
cat events.csv | faery input stdin --dimensions 640x480 \
    filter regularize 30.0 \
    filter render exponential "00:00:00.200000" starry_night \
    output file processed.mp4