Streaming Neural Images
This work addresses critical bottlenecks in INR-based image compression, providing insights for researchers, but it is incremental as it focuses on understanding existing limitations rather than introducing new methods.
The paper tackled the limitations of Implicit Neural Representations (INRs) for image compression, such as computational cost and unstable performance, by analyzing methods like Fourier Feature Networks and Siren, but did not report specific numerical results.
Implicit Neural Representations (INRs) are a novel paradigm for signal representation that have attracted considerable interest for image compression. INRs offer unprecedented advantages in signal resolution and memory efficiency, enabling new possibilities for compression techniques. However, the existing limitations of INRs for image compression have not been sufficiently addressed in the literature. In this work, we explore the critical yet overlooked limiting factors of INRs, such as computational cost, unstable performance, and robustness. Through extensive experiments and empirical analysis, we provide a deeper and more nuanced understanding of implicit neural image compression methods such as Fourier Feature Networks and Siren. Our work also offers valuable insights for future research in this area.