Emergent Source Operator

Authors: Holoscan Team (NVIDIA)
Supported platforms: x86_64, aarch64
Language: C++, Python
Last modified: August 5, 2025
Latest version: 1.0
Minimum Holoscan SDK version: 0.5.0
Tested Holoscan SDK versions: 0.5.0
Contribution metric: Level 1 - Highly Reliable

This operator provides support for Emergent Vision Technologies cameras as video sources. This operator enables high-performance video streaming through Mellanox ConnectX SmartNIC using the Rivermax SDK.

Overview

The EmergentSourceOp is designed to capture video streams from Emergent Vision Technologies cameras with high frame rates and resolution support. It leverages RDMA (Remote Direct Memory Access) capabilities for efficient data transfer and supports various camera parameters for optimal performance. Please refer to Holoscan EVT Setup for more information.

Features

• High Resolution Support: Default resolution of 4200x2160 pixels
• High Frame Rate: Default frame rate of 240 FPS
• RDMA Support: Optional RDMA for enhanced performance
• Configurable Parameters: Adjustable width, height, framerate, exposure, and gain
• Cross-Platform: Supports x86_64 and aarch64 architectures
• Python & C++ APIs: Available in both programming languages

Requirements

• Holoscan SDK: Minimum version 0.5.0 (tested with 0.5.0)
• GXF Extensions: Requires emergent_source extension version 1.0
• Hardware: Mellanox ConnectX SmartNIC for optimal performance
• Camera: Emergent Vision Technologies camera

Parameters

Parameter	Type	Default	Description
signal	gxf::Handle<gxf::Transmitter>	-	Output signal channel
width	uint32_t	4200	Width of the video stream in pixels
height	uint32_t	2160	Height of the video stream in pixels
framerate	uint32_t	240	Frame rate of the video stream in FPS
rdma	bool	false	Enable RDMA for enhanced performance
exposure	uint32_t	3072	Exposure time setting
gain	uint32_t	4095	Analog gain setting

Usage Examples

C++ Example

#include "holoscan/operators/emergent_source/emergent_source.hpp"

// Create the operator
auto emergent_source = fragment.make_operator<holoscan::ops::EmergentSourceOp>(
    "emergent_source",
    holoscan::Arg{"width", 1920},
    holoscan::Arg{"height", 1080},
    holoscan::Arg{"framerate", 60},
    holoscan::Arg{"rdma", true},
    holoscan::Arg{"exposure", 2048},
    holoscan::Arg{"gain", 2048}
);

Python Example

from holoscan.operators import EmergentSourceOp

# Create the operator
emergent_source = EmergentSourceOp(
    fragment,
    width=1920,
    height=1080,
    framerate=60,
    rdma=True,
    exposure=2048,
    gain=2048,
    name="emergent_source"
)

Please refer to High Speed Endoscopy and Laser Detection for requirements and usage examples.