Implicit Neural Representation-Based Continuous Single Image Super Resolution: An Empirical Study
This work addresses a gap in benchmarking and understanding INR-based ASSR for researchers, providing a unified framework and insights into training effects, but it is incremental as it synthesizes and analyzes existing methods rather than introducing a new paradigm.
This paper tackles the lack of empirical analysis in implicit neural representation (INR) methods for arbitrary-scale image super-resolution (ASSR) by conducting a systematic study comparing existing techniques and training configurations, revealing that recent complex INR methods offer only marginal improvements and that model performance is strongly tied to training setups.
Implicit neural representation (INR) has become the standard approach for arbitrary-scale image super-resolution (ASSR). To date, no empirical study has systematically examined the effectiveness of existing methods, nor investigated the effects of different training recipes, such as scaling laws, objective design, and optimization strategies. A rigorous empirical analysis is essential not only for benchmarking performance and revealing true gains but also for establishing the current state of ASSR, identifying saturation limits, and highlighting promising directions. We fill this gap by comparing existing techniques across diverse settings and presenting aggregated performance results on multiple image quality metrics. We contribute a unified framework and code repository to facilitate reproducible comparisons. Furthermore, we investigate the impact of carefully controlled training configurations on perceptual image quality and examine a new loss function that penalizes intensity variations while preserving edges, textures, and finer details during training. We conclude the following key insights that have been previously overlooked: (1) Recent, more complex INR methods provide only marginal improvements over earlier methods. (2) Model performance is strongly correlated to training configurations, a factor overlooked in prior works. (3) The proposed loss enhances texture fidelity across architectures, emphasizing the role of objective design for targeted perceptual gains. (4) Scaling laws apply to INR-based ASSR, confirming predictable gains with increased model complexity and data diversity.