Beamr Imaging Ltd. announced a joint demonstration with dSPACE validating ML-safe video compression for autonomous vehicle data, achieving a 31% file size reduction while preserving machine learning model accuracy.
Key takeaways
Beamr validated ML-safe compression for autonomous vehicle video data in the dSPACE RTMaps ecosystem, preserving machine learning model accuracy.
Testing showed Beamr's CABR compression delivered 31% file size reduction compared to baseline encodes and 97% reduction for uncompressed data.
The demonstration will be presented at the dSPACE user conference on April 21-22, 2026, in Novi, Michigan.
Beamr and dSPACE plan to extend ML-safe compression testing to additional stages including video data simulation and hardware-in-the-loop testing.
Beamr's technology is backed by 53 patents and has won an Emmy Award for Technology and Engineering.