Near-Lossless Deep Feature Compression for Collaborative Intelligence
This addresses the challenge of energy and latency in mobile-cloud AI systems, but it is incremental as it builds on existing compression methods.
The paper tackles the problem of transmitting deep feature data efficiently in collaborative intelligence by proposing a near-lossless compressor, achieving up to 5% bit rate reduction compared to HEVC-Intra.
Collaborative intelligence is a new paradigm for efficient deployment of deep neural networks across the mobile-cloud infrastructure. By dividing the network between the mobile and the cloud, it is possible to distribute the computational workload such that the overall energy and/or latency of the system is minimized. However, this necessitates sending deep feature data from the mobile to the cloud in order to perform inference. In this work, we examine the differences between the deep feature data and natural image data, and propose a simple and effective near-lossless deep feature compressor. The proposed method achieves up to 5% bit rate reduction compared to HEVC-Intra and even more against other popular image codecs. Finally, we suggest an approach for reconstructing the input image from compressed deep features in the cloud, that could serve to supplement the inference performed by the deep model.