Prophesee offers event‑based neuromorphic vision sensors (Metavision) that output asynchronous brightness‑change events instead of full frames, delivering >10 k fps temporal resolution, >120 dB dynamic range, and sub‑10 mW power consumption. The sensors are supplied with a C++/Python SDK, MIPI/USB camera modules and evaluation kits, enabling OEMs to integrate low‑latency, low‑bandwidth visual perception into autonomous robots, drones, XR wearables and defense systems.
Funding
$47.5M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
PVFounders
Product
Problem
Conventional frame‑based cameras generate full image frames at fixed intervals, resulting in high data bandwidth, limited dynamic range, and substantial power consumption. These constraints hinder real‑time perception for applications that require ultra‑fast motion capture, operation under extreme lighting, or energy‑efficient deployment, such as autonomous robotics, XR wearables, and defense systems.
Solution
Prophesee delivers event‑based neuromorphic vision sensors—branded as Metavision—that output asynchronous brightness change events instead of full frames. By processing only pixel‑level changes, the system achieves temporal resolution exceeding 10 000 fps, a dynamic range above 120 dB, and power draw under 10 mW. The offering includes the sensors themselves, a comprehensive Metavision SDK for AI‑driven event processing, evaluation kits for rapid prototyping, and camera modules with standard interfaces (e.g., MIPI). Together, these components enable machines to perceive fast, high‑contrast scenes with minimal data and energy overhead, improving responsiveness and safety in autonomous and immersive applications.
Target Audience
Primary customers are OEMs and system integrators developing autonomous robots, drones, industrial inspection equipment, XR/AR wearables, and defense/aerospace vision subsystems that require high‑speed, low‑power visual sensing.
Features
- Asynchronous event output with >10 k events per second per pixel, providing sub‑millisecond latency
- Wide optical dynamic range (>120 dB) that maintains performance from low‑light to bright‑sun conditions
- Ultra‑low power consumption (<10 mW per sensor) enabling battery‑operated edge devices
- Data reduction of 10‑1000× compared to conventional frame cameras, lowering bandwidth and compute requirements
- Metavision SDK: C++/Python libraries, real‑time event filtering, clustering, and deep‑learning inference pipelines
- Plug‑and‑play evaluation kits and camera modules (MIPI, USB) for rapid integration on PCs, embedded boards, and Raspberry Pi 5
- Built‑in support for eye‑tracking, foveated rendering, and high‑speed SLAM through customizable firmware and IP cores
- Comprehensive documentation, API reference, and optional custom IP or turnkey solution services