Hierarchical Semantic Compression for Consistent Image Semantic Restoration

TL;DR

This paper introduces a hierarchical semantic compression framework that leverages intrinsic semantic spaces from generative models to achieve efficient, consistent image semantic restoration at low bitrates, outperforming existing methods.

Contribution

The paper proposes a novel hierarchical semantic compression framework using a general inversion encoder, feature and semantic compression networks, and an entropy model for improved image semantic restoration.

Findings

Achieves state-of-the-art subjective quality and consistency in image restoration.

Demonstrates superior performance on machine vision tasks with compressed bitstreams.

Provides a new paradigm aligning with human visual understanding.

Abstract

The emerging semantic compression has been receiving increasing research efforts most recently, capable of achieving high fidelity restoration during compression, even at extremely low bitrates. However, existing semantic compression methods typically combine standard pipelines with either pre-defined or high-dimensional semantics, thus suffering from deficiency in compression. To address this issue, we propose a novel hierarchical semantic compression (HSC) framework that purely operates within intrinsic semantic spaces from generative models, which is able to achieve efficient compression for consistent semantic restoration. More specifically, we first analyse the entropy models for the semantic compression, which motivates us to employ a hierarchical architecture based on a newly developed general inversion encoder. Then, we propose the feature compression network (FCN) and semantic compression network (SCN), such that the middle-level semantic feature and core semantics are hierarchically compressed to restore both accuracy and consistency of image semantics, via an entropy model progressively shared by channel-wise context. Experimental results demonstrate that the proposed HSC framework achieves the state-of-the-art performance on subjective quality and consistency for human vision, together with superior performances on machine vision tasks given compressed bitstreams. This essentially coincides with human visual system in understanding images, thus providing a new framework for future image/video compression paradigms. Our code shall be released upon acceptance.