IVAICVLGSep 16, 2024

WaveMixSR-V2: Enhancing Super-resolution with Higher Efficiency

arXiv:2409.10582v32 citationsh-index: 9Has Code
Originality Incremental advance
AI Analysis

This work provides an incremental improvement for super-resolution tasks, benefiting applications requiring efficient high-quality image upscaling.

The paper tackles single image super-resolution by enhancing the WaveMixSR architecture with pixel shuffle operations and a multistage design, achieving state-of-the-art performance on the BSD100 dataset while improving parameter efficiency, latency, and throughput.

Recent advancements in single image super-resolution have been predominantly driven by token mixers and transformer architectures. WaveMixSR utilized the WaveMix architecture, employing a two-dimensional discrete wavelet transform for spatial token mixing, achieving superior performance in super-resolution tasks with remarkable resource efficiency. In this work, we present an enhanced version of the WaveMixSR architecture by (1) replacing the traditional transpose convolution layer with a pixel shuffle operation and (2) implementing a multistage design for higher resolution tasks ($4\times$). Our experiments demonstrate that our enhanced model -- WaveMixSR-V2 -- outperforms other architectures in multiple super-resolution tasks, achieving state-of-the-art for the BSD100 dataset, while also consuming fewer resources, exhibits higher parameter efficiency, lower latency and higher throughput. Our code is available at https://github.com/pranavphoenix/WaveMixSR.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes