Open modules samples

Table of Contents

Open modules C++ samples
Open modules Python samples
HAL C++ samples
Camera Plugins Samples
Samples listed by Module

Open modules C++ samples

C++ samples of the open modules allow you to understand the building blocks of the SDK by starting from basic examples. If you want to start programming from the ground up, you should start with the C++ Get Started tutorial.

In the following table, for each sample, you will find:

its module

if the sample is using the SDK pipelines

what UI framework is used (SDK UI module or OpenCV)

the main algorithms or classes demonstrated in the sample

Name	Description	Module	Using SDK pipelines	UI framework used for display	Main algorithms or classes demonstrated in the sample
metavision_software_info	Prints information on the installed software (e.g. software version)	Base	No	N/A	`SoftwareInfo`
metavision_sdk_get_started	Sample explained step by step that shows how to print statistics and display frames from an event-based camera or from a RAW file	Core	No	SDK UI	`Camera` `PeriodicFrameGenerationAlgorithm`
metavision_active_pixel_detection	Application that detects and masks active pixels (application designed for future sensors)	Driver	No	OpenCV	`Camera` `CDFrameGenerator`
metavision_viewer	Displays data from an event-based camera or from a RAW, DAT or HDF5 file using SDK Driver API	Driver	No	OpenCV	`Camera` `CDFrameGenerator`
metavision_evk3d_viewer	Displays data from an EVK3D stream as a depth map	Driver	No	OpenCV	`Camera`
metavision_file_cutter	Cuts RAW or HDF5 event file from/to given time	Driver	No	N/A	`RAWEventFileWriter` `HDF5EventFileWriter`
metavision_file_info	Gets information about an event file and prints it to console	Driver	No	N/A	`Camera`
metavision_file_to_csv	Converts an event file to CSV-formatted file in C++	Driver	No	N/A	`Camera::from_file`
metavision_file_to_dat	Converts an event file to DAT-formatted file	Driver	No	N/A	`StreamLoggerAlgorithm`
metavision_file_to_hdf5	Converts an event file to an HDF5 event file in C++	Driver	Yes	N/A	`HDF5EventFileWriter`
metavision_file_to_video	Converts an event file to an AVI video	Core	No	N/A	`CvVideoRecorder` `PeriodicFrameGenerationAlgorithm`
metavision_filtering	Filters events (ROI + polarity) and displays them on the screen	Core	Yes	SDK UI	`RoiFilterAlgorithm` `PolarityFilterAlgorithm`
metavision_composed_viewer	Filters events and shows both unfiltered and filtered events	Core	Yes	SDK UI	`FrameCompositionStage`
metavision_csv_viewer	Reads CSV-formatted file and displays events on the screen	Core	Yes	SDK UI	`BaseStage`
metavision_event_frame_generation	Generate and displays histo/diff event frames from event files	Core	No	SDK UI	`EventFrameDiffGenerationAlgorithm` `EventFrameHistoGenerationAlgorithm`
metavision_event_frame_gpu_loading	Preprocesses raw event frame files on a GPU using CUDA	Core	No	N/A
metavision_time_surface	Displays Time Surface of events	Core	No	SDK UI	`MostRecentTimestampBuffer`
metavision_simple_windows	Shows how to create and interact with basic windows	UI	No	SDK UI	`Window` `MTWindow`

Open modules Python samples

Python samples of the open modules allow you to understand the building blocks of the SDK by starting from basic examples. If you want to start programming from the ground up, you should start with the Python Get started tutorial.

In the following table, for each sample, you will find its modules and the main algorithms or classes demonstrated in the sample.

Name	Description	Module	Main algorithms or classes demonstrated in the sample
metavision_sdk_get_started	Sample explained step by step that shows how to print statistics and display frames from an event-based camera or from a RAW file	Core	`EventsIterator` `PeriodicFrameGenerationAlgorithm`
metavision_simple_viewer	Displays events from an event-based camera or from an event file (RAW, DAT or HDF5)	Core	`LiveReplayEventsIterator` `PeriodicFrameGenerationAlgorithm`
metavision_simple_recorder	Record events from an event-based camera to a RAW file	Core	`I_EventsStream`
metavision_file_to_csv	Converts event file (RAW, DAT or HDF5) file to CSV-formatted file in Python	Core	`EventsIterator`
metavision_filtering	Filters events (ROI + polarity) and displays them on the screen	Core	`PolarityFilterAlgorithm` `RoiFilterAlgorithm`
metavision_csv_viewer	Reads CSV-formatted file and displays events on the screen	Core	`PeriodicFrameGenerationAlgorithm`
metavision_interop	Reads multiple AER event file formats and displays events on the screen	Core	`EventsIterator`
metavision_sync	Synchronizes event-based cameras	Core	`initiate_device`
metavision_adaptive_rate	Reads RAW, dynamically splits into event frames based on the content of the stream, and displays result on screen	Core	`AdaptiveRateEventsIterator` `AdaptiveRateEventsSplitterAlgorithm`
metavision_time_surface	Displays Time Surface of events	Core	`MostRecentTimestampBuffer`
demo_event_to_video	Demonstrates Event to Video ML pipeline	Core ML	`EventToVideoLightningModel`
train_event_to_video	Trains Event to Video ML pipeline	Core ML	`EventToVideoLightningModel`
viz_data_event_to_video	Visualize Event to Video training data	Core ML	`GPUEBSIM`
viz_video_to_event_simulator	Demonstrates CPU Video to Event ML pipeline	Core ML	`viz_events`
viz_video_to_event_gpu_simulator	Demonstrates GPU Video to Event ML pipeline	Core ML	`GPUEventSimulator`
demo_corner_detection	Demonstrates Corner Detection ML pipeline	Core ML	`CornerDetectionLightningModel`
train_corner_detection	Trains Corner Detection ML pipeline	Core ML	`CornerDetectionLightningModel`

HAL C++ samples

HAL is the API to access camera functionalities in a generic way (i.e. hiding the hardware specificities of each device). You can discover how to operate and tune sensors and cameras using this API. The samples listed in the table below can be used as starting point.

Note

If you want to build a full-featured application in C++ (with display, data conversion etc.), you might consider using the Driver module that offers a user-friendly API to access the camera and the data with the Camera class (Metavision::Camera). Follow the tutorial Get Started using C++ to discover SDK Driver C++ API. And if you want to code in Python, follow the tutorial Python Get started that shows how you can use EventsIterator to stream events.

Name	Description	Main algorithm or class demonstrated in the sample
metavision_hal_ls	Lists all connected Prophesee devices	`DeviceDiscovery`
metavision_platform_info	Prints information on the platform, connected devices, and installed software (useful for debug)	`I_HW_Identification`
metavision_raw_cutter	Cuts RAW file from/to given time	`I_Decoder` `I_EventsStream`
metavision_hal_seek	Seek in a RAW file	`I_EventsStream`
metavision_hal_showcase	Showcases various HAL API facilities	`I_TriggerIn` `I_TriggerOut` `EventExtTrigger` `I_Monitoring` `I_LL_Biases` `I_EventRateActivityFilterModule` `I_DigitalEventMask` `I_HW_Identification` `I_DigitalCrop`
metavision_hal_sync	Synchronizes event-based cameras	`I_DeviceControl`
metavision_hal_sample_plugin	Provides an example of HAL plugin	`CameraDiscovery`

Note

The HAL API is available in C++ and Python but we currently have samples only for the C++ API. If you want to use Python, we recommend to follow the Python Getting Started guide and check the HAL Python API documentation.

Camera Plugins Samples

Some samples are showing advanced usage of our sensors. When installing the SDK from packages/installer, those samples can be found in : <install-prefix>/share/metavision/hal_psee_plugins/cpp_samples/SAMPLE_NAME.

Name

Description

Sensor

metavision_imx636_facility_casting_sample

Accesses sensor specific implementations of HAL facilities.

For more information, check Hardware Layer Library page.

IMX636

metavision_riscv_logger

Receives log messages from an application embedded in the sensor

For more information, launch metavision_riscv_logger -h.

GenX320