Computationally Efficient Neural Image Compression
This work addresses computational feasibility for practical deployment of neural image compression, but it is incremental as it builds on existing architectures.
The paper tackled the problem of high computational complexity in neural image compression by applying automatic network optimization techniques to a popular architecture, resulting in a reduction of decoder run-time requirements by over 50%.
Image compression using neural networks have reached or exceeded non-neural methods (such as JPEG, WebP, BPG). While these networks are state of the art in ratedistortion performance, computational feasibility of these models remains a challenge. We apply automatic network optimization techniques to reduce the computational complexity of a popular architecture used in neural image compression, analyze the decoder complexity in execution runtime and explore the trade-offs between two distortion metrics, rate-distortion performance and run-time performance to design and research more computationally efficient neural image compression. We find that our method decreases the decoder run-time requirements by over 50% for a stateof-the-art neural architecture.