Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces SEAOTTER, a novel compression framework that integrates a Sensor Embedded Autoencoder with a One-Time Transcode for Efficient Reconstruction, addressing the challenges of high-resolution visual data transmission in robotics. By leveraging a learnable JPEG color and quantization transform, SEAOTTER achieves a remarkable compression ratio of 200:1 while significantly enhancing encoding and decoding speeds compared to existing codecs like AVIF. The framework not only improves ImageNet top-1 accuracy by 8% but also maintains compatibility with established JPEG infrastructure, making it a practical solution for cloud robotics applications.
Achieving a 200:1 compression ratio with 7x faster encoding and 3.5x faster decoding, SEAOTTER redefines efficiency in visual data transmission for robotics.
In robotics systems, vast amounts of visual data are easily captured at high resolution using low-cost, low-power hardware. Yet, limited bandwidth and on-device compute resources prevent full utilization when transmitted via conventional codecs like JPEG/MPEG. Newer codecs, like AV1/AVIF, improve the rate-distortion trade-off, but demand far more resources for encoding, impractical without custom ASICs. Recent asymmetric autoencoders deliver high quality under extreme power and bandwidth constraints, but add prohibitive decoding cost and use bespoke formats that ignore decades of infrastructure built around standards like JPEG. To address these limitations, we introduce a compression framework for cloud robotics based on a Sensor Embedded Autoencoder paired with a One-Time Transcode for Efficient Reconstruction (SEAOTTER). Because the sensor, cloud, and consumer stages face very different power and bandwidth budgets, SEAOTTER combines the compactness of a learned latent with the broad usability of a standard JPEG file. Since naive transcoding degrades performance, we propose a learnable JPEG color and quantization transform that enables increased accuracy for global, dense, and vision-language-based perception. Using SEAOTTER, we train both general-purpose and task-aware transcoding pipelines for a pre-trained, frozen encoder. At a compression ratio of 200:1 and compared to AVIF, we observe 7 times faster encoding, 3.5 times faster decoding, and +8% ImageNet top-1 accuracy, while retaining compatibility with JPEG infrastructure. Our code is available at https://github.com/UT-SysML/seaotter .