Adapting Pretrained Networks for Image Quality Assessment on High Dynamic Range Displays
This work addresses the challenge of limited HDR data for training image quality assessment models, which is important for display technology and content creation, though it is incremental as it builds on existing domain adaptation techniques.
The paper tackled the problem of applying image quality assessment to high dynamic range (HDR) displays by adapting pre-trained networks from standard dynamic range (SDR) data, resulting in models that outperform previous baselines, converge faster, and generalize reliably to HDR inputs.
Conventional image quality metrics (IQMs), such as PSNR and SSIM, are designed for perceptually uniform gamma-encoded pixel values and cannot be directly applied to perceptually non-uniform linear high-dynamic-range (HDR) colors. Similarly, most of the available datasets consist of standard-dynamic-range (SDR) images collected in standard and possibly uncontrolled viewing conditions. Popular pre-trained neural networks are likewise intended for SDR inputs, restricting their direct application to HDR content. On the other hand, training HDR models from scratch is challenging due to limited available HDR data. In this work, we explore more effective approaches for training deep learning-based models for image quality assessment (IQA) on HDR data. We leverage networks pre-trained on SDR data (source domain) and re-target these models to HDR (target domain) with additional fine-tuning and domain adaptation. We validate our methods on the available HDR IQA datasets, demonstrating that models trained with our combined recipe outperform previous baselines, converge much quicker, and reliably generalize to HDR inputs.