SymKL measures signal quality across images audio and 3D data
SymKL: A Unified Representation-Space Measure of Signal Degradation
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Measuring how much a signal gets worse after being changed is important for many technologies, but existing methods only work well for specific types of signals like images or sound. The authors introduce SymKL, a new way to measure degradation that works across different types of signals such as pictures, audio, and 3D objects by comparing how a special model reacts to the original and damaged signals. This method doesn’t need to know how the damage happened and can more accurately track how bad a signal is compared to traditional methods. Experiments show SymKL stays reliable where other methods fail or get stuck.
What this means in practice
- •For audio engineers: Evaluate audio recording quality consistently despite different types of noise or distortions using a unified measure.
- •For 3d graphics developers: Automatically quantify the quality loss in 3D volumetric models after compression or transmission without custom metrics.
Authors
Dhananjaya Jayasundara, Sudarshan Rajagopalan, Vishal M. Patel
Abstract
Quantifying degradation between a pristine signal and its degraded observation is fundamental to signal processing, yet existing full-reference measures are inherently modality-specific. We introduce SymKL, a unified full-reference degradation measure based on implicit neural representations (INRs). An INR is fitted once to the pristine reference and characterized by its diagonal Fisher information, which captures the local information geometry of the learned representation. Evaluating the Fisher information with the pristine reference and a degraded observation yields two parameter-sensitivity profiles whose symmetric KL divergence defines the degradation score. The proposed formulation requires only the reference and degraded observation, without knowledge of the corruption process or degradation severity, making it applicable across images, audio, and 3D data. We further derive an exact closed-form expression for the diagonal Fisher of SIREN, enabling efficient computation. Experiments on image, speech, and volumetric 3D benchmarks demonstrate that SymKL accurately tracks degradation severity, remains monotonic where conventional metrics saturate or fail, and provides a unified modality-agnostic measure of signal degradation. Code will be made publicly available after the review process.