Teams shipping Physical AI run on Deepen.

Leading automotive and robotics teams calibrate, label, and validate their sensor data with us. Real programs, in production, with results they can certify.

Talk to us →Read the AVL story

Case studies.

Outcomes from real programs. We publish a study once a customer story is verified, never before.

Automotive · ADAS

Reliable data for safer ADAS

AVL calibrated its multi-sensor rigs in seconds and labeled large frame counts for the Dynamic Ground Truth reference system, with Deepen Calibrate and Deepen Annotate.

Calibration in secondsMulti-sensor labelingGDPR compliant

Read the full story

Independent, high-precision ground truth reference systems such as AVL DGT play an increasingly important role for our OEMs to objectively develop and validate their vehicles' sensor performance. With Deepen Calibrate, we can ensure that our AVL DGT-sensors are calibrated with high accuracy in a fraction of the time during driving campaigns.

Thomas Guntschnig

Product Portfolio Manager ADAS/AD

AVL

Reliable data for safer ADAS.

AVL's Dynamic Ground Truth (DGT) is a high-resolution reference system that captures a 360 degree view of a vehicle's surroundings from LiDAR, camera, and high-precision GPS. To advance it, AVL needed two things: fast multi-sensor calibration, and high-accuracy labeling delivered quickly. Deepen Calibrate and Deepen Annotate delivered both.

01 / The challenge

Calibration and labeling held the program back

Traditional multi-sensor calibration was slow, expensive, and hard to scale.
Existing annotation tools and providers fell short on experience, custom engineering, and data lifecycle management.
Data integrity sits at the core of DGT, so neither gap was acceptable.

02 / The solution

One stack for calibration and annotation

Deepen Calibrate cut multi-sensor calibration from hours to seconds, across LiDAR, camera, radar, and IMU.
Deepen Annotate delivered efficient, precise labeling on the calibrated data.
A dedicated engineering team handled custom features and GDPR-compliant data operations.

03 / The result

Verified data, captured during driving campaigns

Accurate sensor calibration in seconds, recorded live during driving campaigns.
Precise annotation of large frame counts across many object classes with multiple attributes.
An extremely accurate view of the test environment, with GDPR compliance throughout.

Seconds

Multi-sensor calibration

down from hours

Verified

Reference data

for ADAS and AD validation

GDPR

Compliant data operations

end to end

At scale

Multi-sensor labeling

across object classes

Verified data, at production scale.

The same standards behind every customer program, across 200+ deployments.

98.6%

First-pass

accuracy

200+

Programs

in production

<0.8px

Calibration

accuracy

300K+

Validation

scenarios

See verified results on your data.

Tell us what you are building. We will scope a pilot and prove it on your sensor data.

Talk to us →Explore the stack