PerQ: Inverse Generative Modeling for Neural Image Compression via Quantization Error Compensation
Neural image codecs face a fundamental trade-off between pixel fidelity and perceptual quality. Distortion-optimized codecs preserve pixel-wise fidelity but produce blurred reconstructions at low bitrates, whereas generative codecs leverage learned priors to improve perceptual quality but tend to…
arXiv - cs.CV (CV)2 min read