Coarse-to-Fine: Progressive Image Compression for Semantically Hierarchical Classification

2026-05-07

AI Summary

This work proposes a semantic hierarchy-aware progressive image compression framework that explicitly incorporates semantic structure into the compression process. While existing methods primarily focus on sample-level difficulty adaptation, they often fail to preserve high-level semantic information effectively at low bitrates. To address this limitation, the proposed approach leverages CLIP embeddings to construct a semantic hierarchy over ImageNet-1K classes and explicitly decomposes and optimizes corresponding channel blocks in the latent representation. This enables semantically scalable transmission—from coarse- to fine-grained semantics—within a single bitstream. Experimental results demonstrate that the method significantly improves coarse-grained recognition performance at low bitrates while maintaining fine-grained classification accuracy at higher bitrates, outperforming current state-of-the-art progressive codecs across the bitrate spectrum.

Abstract

Recent advances in learned image compression (LIC) have enabled practical deployments, spurring active research into image compression for machines and progressive coding schemes. However, their integration remains under-explored: prior works on progressive machine codec predominantly target sample-level difficulty adaptation (i.e., easy-to-hard), without considering semantic-level scalability. In this work, we introduce a semantic hierarchy-aware progressive codec that enables semantic scalability (i.e., coarse-to-fine) from a single bitstream. We first systematically categorize ImageNet-1K classes into CLIP embedding-based semantic hierarchies. Based on a channel-wise autoregressive framework, we decompose latent representations into hierarchically ordered channel blocks, each explicitly optimized for a corresponding semantic hierarchy. Extensive experiments demonstrate that our approach substantially improves coarse-level recognition at low bitrates while maintaining fine-grained accuracy at higher bitrates. By reframing progressive transmission through the lens of semantic scalability, our work provides an efficient and interpretable solution for task-adaptive image coding, outperforming existing progressive codecs under hierarchical evaluation.

Problem

• semantic scalability
• progressive image compression
• learned image compression
• hierarchical classification
• coarse-to-fine

Innovation

• semantic scalability
• progressive image compression
• learned image compression
• coarse-to-fine
• hierarchical classification