Event camera
Imaging sensor that reports brightness changes asynchronously
An event camera is an image sensor that detects local brightness shifts instead of capturing full frames. Each pixel works independently to output data only when illumination changes, enabling high speed and low power use.
- Pixels operate independently and asynchronously
- Reports changes in brightness with microsecond resolution
- Offers 120 dB dynamic range with less motion blur
- Suitable for autonomous vehicles and robotics
- Early models had 100 pixel resolution in 2014
What it is
An event camera, also called a neuromorphic camera or silicon retina, is a type of imaging sensor. Unlike standard cameras that capture full images at fixed intervals using a shutter, this device responds to local changes in brightness. Each pixel operates on its own and asynchronously. It remains silent until it detects a shift in light intensity. When a change exceeds a set threshold, the pixel generates an event. This event is a small data packet containing the pixel address and a precise timestamp. The output is a continuous stream of these events rather than a sequence of frames. This design allows for microsecond temporal resolution and a dynamic range of 120 dB. It reduces motion blur and exposure issues common in traditional sensors.
How it came about
The concept of spatial event-driven convolution was postulated in 1999, prior to the invention of the Dynamic Vision Sensor (DVS). The DVS was later developed during the EU project CAVIAR. Early models in 2014 had a resolution limited to 100 pixels. By 2019, newer entries achieved a resolution of 640 by 480 pixels. In 2020, the term retinomorphic was adopted for a specific sensor design using a resistor and photosensitive capacitor. These retinomorphic sensors have so far only been studied in research environments. The technology continues to evolve with various types such as sDVS and CSDVS being introduced to improve sensitivity and reduce redundancy.
Inside the chip
Each pixel in an event camera stores a reference brightness level. It continuously compares this reference to the current light level. If the difference passes a threshold, the pixel resets its reference and sends an event. This event may include polarity, indicating if brightness increased or decreased. Some models also report instantaneous illumination. Temporal contrast sensors like DVS produce events with polarity, while temporal image sensors report intensity. The DAVIS model combines a global shutter active pixel sensor with a DVS, allowing it to produce both frames and events. CSDVS adds a resistive network to filter spatial redundancy. Many units also include an inertial measurement unit. Retinomorphic sensors use a capacitor that changes capacitance under light, generating voltage only when intensity shifts.
How it was made
Resolution has improved from 100 pixels in early versions to 640x480 in later models. The sensors are designed to integrate with asynchronous computing architectures. The actual temporal resolution varies from microseconds to milliseconds depending on lighting and sensor design. No specific packaging or power draw figures are provided for the general category.
Where it went
These sensors are suited for machine vision tasks such as object recognition, robotics, and autonomous vehicles. Their low power consumption and reduced heat generation make them attractive for military applications, including infrared sensing. They are also considered for space imaging, security, defense, and industrial monitoring. The ability to handle scenes with mixed bright and dim regions without averaging is valuable where stabilizing the line of sight is difficult. Research into color sensing is ongoing, but current models are not yet convenient for applications requiring color data. They excel in tracking optical flow and coping with high-speed movement.
Versions and family
Several types exist, including Dynamic Vision Sensors (DVS) and sensitive-DVS (sDVS). The DAVIS sensor combines event detection with standard frame capture. Center Surround Dynamic Vision Sensors (CSDVS) use a network to reduce output redundancy. Retinomorphic sensors represent another class, using a resistor and photosensitive capacitor series circuit. These differ from standard designs by producing signals inherently through capacitance changes. Image reconstruction methods vary, using temporal smoothing or optimization techniques. Algorithms for motion detection must account for the lack of texture and color in the event stream, often employing motion-compensation models.
Why it matters
Event cameras offer significant advantages over conventional sensors in high-speed and high-dynamic-range scenarios. They enable accurate tracking of object and camera movement with minimal motion blur. The technology supports the development of neuromorphic computing systems by providing asynchronous data streams. Potential applications span from autonomous driving to space exploration. The US military is exploring their use due to efficiency benefits. While color sensing remains a challenge, the core technology provides a robust solution for environments where traditional cameras struggle with exposure and latency. The evolution from low-resolution prototypes to higher-density arrays marks steady progress.


Questions people ask about Event camera
How does an event camera differ from a standard camera?
Standard cameras capture full frames at fixed intervals using a shutter. Event cameras do not use shutters. Instead, each pixel works independently to report only when brightness changes. This creates an asynchronous stream of data rather than a sequence of images.
What are the main advantages of this sensor type?
They offer microsecond temporal resolution and a 120 dB dynamic range. This results in less motion blur and fewer exposure issues. They also consume less power and generate less heat, making them ideal for high-speed or battery-powered applications.
Can event cameras capture color images?
Current event cameras primarily yield grey-scale information. Research into color sensing is underway, but the technology is not yet convenient for applications that require color data. Most models focus on detecting brightness changes rather than spectral information.
What is a retinomorphic sensor?
It is a specific type of event sensor named in 2020. It uses a resistor and a photosensitive capacitor in series. The capacitor changes capacitance under light, generating a voltage only when intensity changes. This differs from other sensors that use photodiodes and circuits.
Good reading and sources
The facts on this page come from the Wikipedia article "Event camera", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.
Like Event camera
All image and other sensors
WorldHazcam
Wide-angle engineering cameras for rover navigationSensor
🇺🇸 United StatesCachecam
Sample verification camera for the Perseverance Mars roverSensor