Ouster's Rev8 brings native colour to Lidar

In what Ouster describes as the most significant advancement in Lidar's history, it has recently unveiled Rev8, a new family of digital Lidar sensors. Powered by its next-generation L4 Ouster Silicon, the company claims this is the world's first native colour Lidar.

At the heart of the announcement is a fundamental shift in how colour and depth data are acquired. Rather than fusing camera imagery and point cloud geometry in post-processing, Rev8 captures both in a single photon event. Every point is generated with 48-bit colour and 3D geometry simultaneously, eliminating the calibration overhead and temporal misalignment that has long complicated multi-sensor workflows in mapping and SLAM applications.

"Rev8 is the most advanced family of Lidar sensors ever released and sets a new standard in sensing," said Ouster CEO Angus Pacala. "With the L4 Ouster Silicon, we are delivering on the promise of our digital architecture to deliver exponential improvements in performance, doubling our core specs and simultaneously introducing the world's first native colour Lidar to give machines 3D human-like sight for the next era of Physical AI. Rev8 is the foundational technology that will allow customers to move from prototype to commercial production at scale, providing the reliability and affordability required to enable real-world autonomy across industries."

Single pass mapping with a single sensor

For geospatial professionals, the implications for mobile mapping and reality capture are considerable. SLAM workflows that previously required separate Lidar and camera passes, followed by time-consuming fusion and drift correction in post-processing, can now be executed in a single pass with a single sensor. Ouster demonstrated this capability using the OS1 Max in Lofoten, Norway, where the resulting colourized point cloud, captured in challenging foggy conditions, so closely resembled a photograph that the company used it to highlight the sensor's colour fidelity and spatial accuracy.

Another example of the capabilities of the new family of Lidar sensors is the Cathedral of Seville, one of the largest Gothic cathedrals in the world, spanning 11,520 square meters of vaulted ceilings, stained glass, and centuries of intricate stonework. Documenting a structure of this scale traditionally requires dozens of scanner setup positions across multiple days, followed by hours of registration and texturing in post-production. The Rev8 OS1 Max completes the same task in a single afternoon walkthrough. The sensor records native colour and depth simultaneously in the same photon event, with 48-bit colour locked to its geometry at the point level, eliminating the need for separate texturing workflows. The outcome is a complete, richly detailed capture of one of the world's most complex interiors, delivered in a fraction of the time.

The performance specifications underpinning Rev8 are notable. The L4 Silicon processes up to 10.4 million points per second and pushes 22.4 gigabits per second of data off-chip, with a 40kHz measurement rate and picosecond timing precision. The flagship 256-channel OS1 Max offers sensing up to 500 metres with a 45° field of view – double the resolution of the Rev7 OS2, in a form factor a quarter of the size. Dynamic range reaches 116dB, with sensitivity spanning 1 lux to 2 million lux, addressing the lighting extremes that routinely challenge survey-grade capture in the field.

Native colour fidelity

Colour science is handled through a partnership with Fujifilm, whose imaging expertise is embedded directly into the L4 architecture via hardware-enabled high dynamic range. The result, Ouster says, is colour data with the fidelity required for automated feature extraction, classification and training of Physical AI world models, without reliance on a co-registered camera.

The Rev8 OS family spans upgraded versions of the OS0, OS1, and OSDome sensors alongside the new OS1 Max, designed with functional safety, scalability and affordability in mind. Ouster positions the lineup as suitable for both data collection and production deployment on the same sensor hardware, a consideration likely to appeal to operators managing large-scale mapping programmes who have historically needed separate sensor configurations for each phase.

The announcement positions Ouster squarely in the emerging Physical AI market, where high-quality, georeferenced 3D colour data is increasingly seen as critical infrastructure for training autonomous systems across automotive, robotics, industrial and smart infrastructure applications.