SBX Robotics

Get 1000x more training data for computer vision.

Skip manual labeling & iterate quickly.
Multiply 25 real-world images into 25,000 synthetic training images for deep learning.

"Our model trained on SBX data significantly outperformed one trained on data we collected."
- Tarik Kelestemur, Researcher
Institute for Experiential Robotics at Northeastern University
"The ability of models trained on SBX data to generalize on fairly diverse items is impressive."
- Marek Cygan, CTO
NoMagic company logo
"SBX kickstarts projects far faster than real-world data acquisition and labeling."
- James Servos, Perception Team Manager
OTTO Motors
"SBX showed us synth data is viable for our AI needs. We look forward to working together."
- Daniel Grollman, Lead R&D Engineer, PhD
Plus One Robotics

Synthetic training data is the fastest and cheapest way to improve or bootstrap a computer vision model. 

Skip costly hardware setup, data collection, data annotation, and data cleaning. Using technology from film and gaming, we produce realistic, perfectly labeled training datasets for object detection, segmentation, and 6D pose estimation models.

Each SBX dataset is the product of iterative testing and optimization to achieve the best performance on real-world data.

Try it yourself with our free 10K sample dataset & tutorial.

What is synthetic training data?

RGB image of a robot picking up a box
Segmentation annotations for image of a robot picking up a box
Depth data for image of a robot picking up a box

Case Studies

Detecting forklifts with OTTO Motors

Autonomous forklifts need to keep track of their human-driven counterparts to operate with them side-by-side. The team at OTTO Motors looked to SBX synthetic data to bootstrap and improve computer vision systems in their line of AMRs.

"SBX kickstarts projects far faster than real-world data acquisition and labeling."

- James Servos, Perception Lead

OTTO Motors

Thanks to a model trained on SBX synthetic data, the ER-FLEX robot can now fetch part bins in a manufacturing facility.

To pick a bin from the shelf, the robot must understand its precise location and orientation (6D pose). Synthetic data offers an easy alternative to the nearly impossible task of collecting this data in the real world.

Enabled Robotics

Item manipulation with Enabled Robotics

Building a smart cafeteria

Our client was building a smart camera to identify table tops in a cafeteria, some of which may still have items on them.

Setting up sensors to collect data across different cafeterias was a major hurdle to scale this client's technology from the first prototype to other locations.

With our synthetic data generator, SBX was able to produce an infinite variety of virtual cafeterias to ensure that our client's vision system adapted to new environments.

Other Applications

RGB image of grocery items in a binSegmentation data for image of grocery items in a binRGB image of items in a manufacturing facilityRGB image of boxes in a storage facilityRGB image of a robot in a logisitcs facility6D pose annotation for image of parts in a manufacturing facilitySegmentation annotations for image of boxes in a storage facilitySegmentation annotation for image of robot in a logistics facility

E-grocery Fulfillment




Founding team

LinkedInIan Dewancker photo

Ian Dewancker, CEO

Led applied research projects at UberATG, Kindred AI, and SigOpt focusing on computer vision, robotics, and optimization for machine learning.

LinkedInJoshua Kuntz photo

Joshua Kuntz, CTO

Built the tech and team behind merchant marketplace, growing it to 100k+ active merchants and 45 engineers, PMs, and designers.

1000x my data, please!

Share 25 images from your vision system, and we will generate an optimized training set of 25,000 annotated synthetic images.

Not sure? Try our synthetic data tutorial with a free 10,000 image dataset, tutorial videos, and open source code to train a model.
Max file size 10MB.
Upload failed. Max size for files is 10 MB.
Multiplier engine engaged. Our flight crew will be in touch shortly.
Oops! Something went wrong with the multiplier machine..