An Autoencoder-based Learned Image Compressor: Description of Challenge Proposal by NCTU
This is an incremental improvement for image compression tasks, targeting participants in the CLIC challenge.
The authors tackled the problem of lossy image compression by proposing an autoencoder-based system for the CLIC 2018 challenge, aiming to achieve good subjective quality under a 0.15 bits-per-pixel constraint.
We propose a lossy image compression system using the deep-learning autoencoder structure to participate in the Challenge on Learned Image Compression (CLIC) 2018. Our autoencoder uses the residual blocks with skip connections to reduce the correlation among image pixels and condense the input image into a set of feature maps, a compact representation of the original image. The bit allocation and bitrate control are implemented by using the importance maps and quantizer. The importance maps are generated by a separate neural net in the encoder. The autoencoder and the importance net are trained jointly based on minimizing a weighted sum of mean squared error, MS-SSIM, and a rate estimate. Our aim is to produce reconstructed images with good subjective quality subject to the 0.15 bits-per-pixel constraint.