Beamr Imaging Ltd. (NASDAQ: BMR) announced a new service called Beamr Blueprint, designed to help autonomous vehicle teams identify where video compression can be safely applied in their data pipelines, according to a company statement.

The service involves Beamr's video data experts mapping an autonomous vehicle team's pipeline from in-vehicle capture through cloud storage to model training, then testing compression against the team's own models and data. The engagement produces a verified plan indicating where lossy or lossless compression can reduce storage and networking costs without affecting model accuracy.

Beamr said a global autonomous vehicle program has already begun a Blueprint engagement, though the company did not identify the client.

The company's Content-Adaptive Bitrate technology, known as CABR, has delivered up to 50% reductions in storage, networking, and compute in published benchmarks. For object detection tasks, CABR showed under 2% difference in mean Average Precision, which Beamr describes as within a model's expected variance.

"Tier-1 suppliers and OEMs have asked us to bring our expertise directly into their pipelines, with the judgment to know where compression is safe and where it is not," said Sharon Carmel, Beamr CEO. "Blueprint AV turns that work into a defined offering, and paired with our ecosystem partnerships, it gives teams a faster path from evaluation to adoption - while opening a commercial channel for Beamr alongside our technology licensing."

Beamr also noted that earlier in July its technology became available in the RTMaps AI Store of Intempora, a dSPACE company. The Blueprint service will also be offered through partners.