Beamr Launches ML-Safe Lossless Compression for Autonomous Vehicle Fleets
Beamr adds GPU-accelerated, edge-based lossless compression to its ML-Safe suite so AV teams can store more raw sensor video without new hardware.
Rhea-AI Summary
Beamr Imaging (BMR) has launched ML-Safe lossless video compression for autonomous vehicle fleets, reducing 12-bit Bayer footage file sizes by 47%.
The ML-Safe solution now combines model-validated lossy compression with bit-exact lossless compression, aimed at preserving machine-learning accuracy while cutting data volume. Beamr’s lossless technology runs on the existing video encoder integrated into the GPU in AV data loggers, so no new hardware is required. The company reports real-time compression of multiple camera streams and nearly doubled data-capture capacity on test vehicles. Beamr will demonstrate the solution at AutoSens Europe in Barcelona on September 22–24 at stand 131.
Positive
- None.
Negative
- None.
Key Figures
- File-size reduction
- 47%
- 12-bit Bayer recordings from 8-megapixel cameras
- Camera support
- 16 cameras
- Maximum cameras reported for a single test vehicle
- Typical camera count
- 11 cameras
- Typical single test vehicle
- Raw image format
- 12-bit Bayer
- Camera sensor raw output
- Camera resolution
- 8-megapixel
- Recordings used in Beamr's testing
Historical Context
-
Reported first structured global AV engagement alongside ML-safe compression and first-half results.
24h Move is the share-price change in the day after each event; other market factors may also have contributed.
Key Terms
lossless compression technical
bit-exact technical
gpu technical
AI-generated analysis. How Rhea-AI works. Not financial advice.
Autonomous vehicle test fleets generate more video than they can store or move. Beamr’s lossless compression keeps footage bit-exact at capture in the test vehicles, while reducing file sizes by
Join the demonstration at AutoSens Europe, Barcelona, September 22–24, stand 131
Herzliya, Israel, Sept. 16, 2026 (GLOBE NEWSWIRE) -- Beamr Imaging Ltd. (Nasdaq: BMR), a leader in video optimization technology and solutions, today launched lossless compression as part of ML-Safe video data compression for autonomous vehicles (AV). The ML-Safe solution, enabling AV teams to reduce video data file sizes while preserving ML model accuracy, now covers lossy compression that is confirmed against the model's own accuracy metrics, and lossless compression that is bit-exact by definition.
Beamr will demonstrate its ML-Safe solution at AutoSens Europe in Barcelona, September 22–24, stand 131.
AV test fleets generate more video than they can store or move. A single test vehicle runs as many as 16 cameras, and 11 is typical. They can record tens of terabytes in one shift. Cameras commonly produce 12-bit Bayer, the camera sensor's raw output before it enters the AV pipeline. Until now AV teams often had to keep the footage raw, without compression, since the encoder in the test vehicle did not accept the 12-bit Bayer format. As a result, AV teams had two choices: keep the raw footage and pay for it in drives, swaps and upload time, or delete video data and hope a future model never needs it.
Beamr's lossless compression addresses the AV teams need by taking the 12-bit Bayer and compressing it with zero loss. In addition, data-capture capacity is almost doubled, data is preserved, and no new hardware is needed. The work runs on the video encoder that is integrated in the GPU already in the data logger. Since the encoder runs separately from the GPU compute cores, compression draws on capacity the AI models are not using.
Beamr's lossless compression is designed to compress multiple cameras in real time on the video encoder of a single GPU. In Beamr's testing on 12-bit Bayer recordings from 8-megapixel cameras, compressed files were
“Through our joint work with AV and ADAS companies we identified their need for lossless compression at the edge, in the data-collection vehicle, supporting multiple camera streams in real time. So we built it,” said Sharon Carmel, Beamr CEO. “By running Beamr's lossless solution accelerated by GPUs, we are demonstrating a solution that reduces the data at the point of capture while preserving every bit of the original camera data.”
"Lossless compression was the piece missing from Beamr’s ML-Safe video data compression solution for AV,” said Dani Megrelishvili, Beamr CPO. “We are already offering ML-Safe lossy compression tested against the team's own models. Now teams that need every bit kept can also compress in the test vehicle. This puts Beamr at the point where data is created, and what is saved there is saved again at every step that follows: on the drives, on the way to the cloud, and in every training run that reads it."
Autonomous vehicle and ADAS teams can test Beamr ML-Safe compression on their own video data and pipeline design. To book a meeting at AutoSens Europe, Barcelona (stand 131), or request lossless evaluation on your data, visit beamr.com/lossless.
About Beamr
Beamr (Nasdaq: BMR) is a world leader in content-adaptive video compression, trusted by top media companies including Netflix and Paramount. Beamr’s perceptual optimization technology (CABR) is backed by 53 patents and a winner of Emmy® Award for Technology and Engineering. The innovative technology reduces video file sizes by up to
Beamr powers efficient video workflows across high-growth markets, such as media and entertainment, user-generated content, machine learning, and autonomous vehicles. Its flexible deployment options include on-premises, private or public cloud, with convenient availability for Amazon Web Services (AWS) and Oracle Cloud Infrastructure (OCI) customers.
For more details, please visit www.beamr.com or the investors’ website www.investors.beamr.com and follow us on Linkedin and X.
Forward-Looking Statements
This press release contains “forward-looking statements” that are subject to substantial risks and uncertainties. Forward-looking statements in this communication may include, among other things, statements about Beamr’s strategic and business plans, technology, relationships, objectives and expectations for its business, the impact of trends on and interest in its business, intellectual property or product and its future results, operations and financial performance and condition, including statements relating to the capabilities, performance, applications, market opportunities, adoption, commercialization and potential benefits of the Company’s technology, products and solutions, and the Company’s ability to execute its growth strategy and create commercial opportunities. All statements, other than statements of historical fact, contained in this press release are forward-looking statements. Forward-looking statements contained in this press release may be identified by the use of words such as “anticipate,” “believe,” “contemplate,” “could,” “estimate,” “expect,” “intend,” “seek,” “may,” “might,” “plan,” “potential,” “predict,” “project,” “target,” “aim,” “should,” “will,” “would,” or the negative of these words or other similar expressions, although not all forward-looking statements contain these words. Forward-looking statements are based on the Company’s current expectations and are subject to inherent uncertainties, risks and assumptions that are difficult to predict. Further, certain forward-looking statements are based on assumptions as to future events that may not prove to be accurate. For a more detailed description of the risks and uncertainties affecting the Company, reference is made to the Company’s reports filed from time to time with the Securities and Exchange Commission (“SEC”), including, but not limited to, the risks detailed in the Company’s annual report filed with the SEC on February 26, 2026 and in subsequent filings with the SEC. Forward-looking statements contained in this announcement are made as of the date hereof and the Company undertakes no duty to update such information except as required under applicable law.
Investor Contact:
investorrelations@beamr.com
FAQ
AI-generated questions and answers. How Rhea-AI works. Not financial advice.
What type of video data does Beamr’s new lossless compression support?
Beamr’s lossless compression targets 12-bit Bayer recordings, which are the raw outputs from autonomous vehicle camera sensors before entering the AV processing pipeline.
How is the compression implemented on autonomous vehicle test vehicles?
The work runs on the video encoder integrated into the GPU already present in the data logger. Since the encoder is separate from the GPU compute cores, it uses capacity that AV AI models are not using, and is designed to compress multiple cameras in real time on a single GPU.
What compression results did Beamr report in its testing?
In testing on 12-bit Bayer recordings from 8-megapixel cameras, Beamr reports that compressed files were 47% smaller than the raw footage, while remaining bit-exact.
How can AV or ADAS teams evaluate Beamr’s ML-Safe compression?
Autonomous vehicle and ADAS teams can test Beamr ML-Safe compression on their own video data and pipeline design. To book a meeting at AutoSens Europe in Barcelona (stand 131) or request a lossless evaluation on their data, they are directed to visit beamr.com/lossless.