Prophesee (Formerly Chronocam) is the inventor and supplier of 4 Event-Based sensors generations, including commercial-grade versions as well as industry’s largest software suite. The company focuses on Industrial, Mobile-IoT and Automotive applications.
SLAMcore develops Localisation and mapping solutions for AR/VR, robotics & autonomous vehicles.
CelePixel (formerly Hillhouse Technology) offer integrated sensory platforms that incorporate various components and technologies, including a processing chipset and an image sensor (a dynamic vision sensor called CeleX).
Serrano-Gotarredona, R., Oster, M., Lichtsteiner, P., Linares-Barranco, A., Paz-Vicente, R., Gomez-Rodriguez, F., Riis, H.K., Delbruck, T., Liu, S.-H., Zahnd, S., Whatley, A.M., Douglas, R., Hafliger, P., Jimenez-Moreno, G., Civit, A., Serrano-Gotarredona, T., Acosta-Jimenez, A., Linares-Barranco, B., AER building blocks for multi-layer multi-chip neuromorphic vision systems,
Advances in neural information processing systems, 1217-1224, 2006.
Kirkland, P., Di Caterina, G., Soraghan, J., Matich, G., Neuromorphic technologies for defence and security,
SPIE vol 11540, Emerging Imaging and Sensing Technologies for Security and Defence V; and Advanced Manufacturing Technologies for Micro- and Nanosystems in Security and Defence III; 2020.
Sabater, A., Montesano, L., Murillo, A., Event Transformer. A sparse-aware solution for efficient event data processing,
IEEE Conf. Computer Vision and Pattern Recognition Workshops (CVPRW), 2022. PDF, Supp. Video, Code.
Lee, A. J., Cho, Y., Yoon, S., Shin, Y., Kim, A., ViViD: Vision for Visibility Dataset,
IEEE Int. Conf. Robotics and Automation (ICRA) Workshop: Dataset Generation and Benchmarking of SLAM Algorithms for Robotics and VR/AR, 2019.
MNIST-DVS and FLASH-MNIST-DVS datasets are based on the original frame-based MNIST dataset. MNIST-DVS are DVS128 recordings of moving MNIST digits (at 3 scales), while FLASH-MNIST-DVS datasets are recorded by flashing the digits on a monitor.
POKER-DVS. From a set of DVS recordings of very fast poker card browsing, 32x32 pixel windows tracking the symbols are cropped. On average each symbol lasts about 10-30ms.
SLOW-POKER-DVS. Paper printed poker card symbols are moved at "human speed" in front of a DVS camera and recorded at 128x128 resolution.
Sync Toolbox. This open-source toolbox provides a QT-based GUI to allow easy access for hardware-level multi-sensor synchronization (Prophesee Gen 3.1 included and tested). After proper configuration of the software, users can seamlessly record new ROS bags.
BIMVEE Python tools for Batch Import, Manipulation, Visualisation and Export of Events and other timestamped data. Imports from various file formats into a common workspace format, including native Python import of rosbags.
Tonic provides publicly available event datasets and data transformations much like Torchvision/audio.
dv_ros ROS package for accumulating event frames with iniVation Dynamic Vision System's dv-sdk.
dvs_event_server ROS package used to transport "dvs/events" ROS topic to Python through protobuf and zmq, because Python ROS callback has a large delay.
AEStream A fast C++ library with a Python interface for streaming Address Event representations directly from Inivation and Prophesee cameras to various sources, such as STDOUT, UDP (network), or PyTorch.
AEDAT decoder A fast AEDAT 4 Python reader, with a Rust underlying implementation.
aedat-rs Standalone Rust library for decoding AEDAT 4 files for use in bespoke Rust event systems.
expelliarmus A pip-installable Python library to decode DAT, EVT2 and EVT3 files generated by Prophesee cameras to structured NumPy arrays.