EVC: Towards Real-Time Neural Image Compression with Mask Decay
This work addresses the challenge of making neural image compression practical for real-time applications by reducing complexity and enabling variable-bit-rate support, though it appears incremental as it builds on existing neural compression methods.
The paper tackles the problem of high complexity and separate models for different rate-distortion trade-offs in neural image compression by proposing EVC, an efficient single-model variable-bit-rate codec that achieves real-time performance (30 FPS) with 768x512 images and outperforms VVC in rate-distortion performance, with a small model reaching 30 FPS for 1920x1080 images.
Neural image compression has surpassed state-of-the-art traditional codecs (H.266/VVC) for rate-distortion (RD) performance, but suffers from large complexity and separate models for different rate-distortion trade-offs. In this paper, we propose an Efficient single-model Variable-bit-rate Codec (EVC), which is able to run at 30 FPS with 768x512 input images and still outperforms VVC for the RD performance. By further reducing both encoder and decoder complexities, our small model even achieves 30 FPS with 1920x1080 input images. To bridge the performance gap between our different capacities models, we meticulously design the mask decay, which transforms the large model's parameters into the small model automatically. And a novel sparsity regularization loss is proposed to mitigate shortcomings of $L_p$ regularization. Our algorithm significantly narrows the performance gap by 50% and 30% for our medium and small models, respectively. At last, we advocate the scalable encoder for neural image compression. The encoding complexity is dynamic to meet different latency requirements. We propose decaying the large encoder multiple times to reduce the residual representation progressively. Both mask decay and residual representation learning greatly improve the RD performance of our scalable encoder. Our code is at https://github.com/microsoft/DCVC.