UniComp: Rethinking Video Compression Through Informational Uniqueness
This work addresses video compression efficiency for applications requiring high-quality visual data under computational limits, representing an incremental improvement over prior methods.
The paper tackles video compression by introducing an information uniqueness framework to maximize information fidelity under computational constraints, and demonstrates that UniComp outperforms existing methods in preserving visual tokens with limited resources.
Distinct from attention-based compression methods, this paper presents an information uniqueness driven video compression framework, termed UniComp, which aims to maximize the information fidelity of video representations under constrained computational budgets. Starting from the information-theoretic perspective, we formulate the vision compression as an optimization problem that minimizes conditional entropy (reconstruction error) between retained and full tokens. To achieve this, we introduce the notion of information uniqueness to measure intrinsic redundancy among tokens to link with reconstruction error. Based on uniqueness, we design three modules-Frame Group Fusion, Token Allocation, and Spatial Dynamic Compression-that progressively perform semantic frame grouping, adaptive resource allocation, and fine-grained spatial compression. Extensive experiments demonstrate that UniComp consistently outperforms existing compression methods in preserving essential visual tokens under limited computational budgets, highlighting the pivotal role of information uniqueness in token compression efficacy.