Ultra High Fidelity Image Compression with ℓ_∞-constrained Encoding and Deep Decoding
In many professional fields, such as medicine, remote sensing and sciences, users often demand image compression methods to be mathematically lossless. But lossless image coding has a rather low compression ratio (around 2:1 for natural images). The only known technique to achieve significant compression while meeting the stringent fidelity requirements is the methodology of ℓ_∞-constrained coding that was developed and standardized in nineties. We make a major progress in ℓ_∞-constrained image coding after two decades, by developing a novel CNN-based soft ℓ_∞-constrained decoding method. The new method repairs compression defects by using a restoration CNN called the ℓ_∞ to map a conventionally decoded image to the latent image. A unique strength of the ℓ_∞ is its ability to enforce a tight error bound on a per pixel basis. As such, no small distinctive structures of the original image can be dropped or distorted, even if they are statistical outliers that are otherwise sacrificed by mainstream CNN restoration methods. More importantly, this research ushers in a new image compression system of ℓ_∞-constrained encoding and deep soft decoding (ℓ_∞^2). The ℓ_∞^2 approach beats the best of existing lossy image compression methods (e.g., BPG, WebP, etc.) not only in ℓ_∞ but also in ℓ_2 error metric and perceptual quality, for bit rates near the threshold of perceptually transparent reconstruction. Operationally, the new compression system is practical, with a low-complexity real-time encoder and a cascade decoder consisting of a fast initial decoder and an optional CNN soft decoder.
READ FULL TEXT