Synthera launches Chameleon synthetic-data platform
Synthera introduced Chameleon, a no-code Windows platform for generating and governing annotated synthetic training data for computer-vision models.
Evidence notes
- Chameleon generates unlimited annotated images on local Windows infrastructure, with outputs including segmentation masks, bounding boxes and keypoints.
- The platform combines data generation with evaluation, privacy controls, provenance and versioning so teams can manage a repeatable vision-model training workflow.
- Synthera positioned the system for cameras, robots, autonomous vehicles and safety applications where real-world data is scarce, sensitive or expensive to label.
Company context
Synthera builds synthetic training-data infrastructure for computer-vision systems used in cameras, robots, vehicles and safety applications. Founded in 2025 around the acquisition of Mindtech, it uses the acquired Chameleon platform as the core of an analysis-led pipeline for creating controlled, annotated datasets when real imagery is scarce, restricted or unsafe to capture. The company says the technology has delivered more than one million training images.