Universal Efficient Variable-rate Neural Image Compression
This addresses practical deployment issues for learning-based image compression systems, representing an incremental improvement.
The paper tackles computational complexity and rate flexibility challenges in learning-based image compression by proposing two universal modules (Energy-based Channel Gating and Bit-rate Modulator) that reduce FLOPs by over 50% in convolution layers and enable arbitrary bit-rate output with a single model.
Recently, Learning-based image compression has reached comparable performance with traditional image codecs(such as JPEG, BPG, WebP). However, computational complexity and rate flexibility are still two major challenges for its practical deployment. To tackle these problems, this paper proposes two universal modules named Energy-based Channel Gating(ECG) and Bit-rate Modulator(BM), which can be directly embedded into existing end-to-end image compression models. ECG uses dynamic pruning to reduce FLOPs for more than 50\% in convolution layers, and a BM pair can modulate the latent representation to control the bit-rate in a channel-wise manner. By implementing these two modules, existing learning-based image codecs can obtain ability to output arbitrary bit-rate with a single model and reduced computation.