用新型网络实现图像补全高速推理,速度比现有方法快10倍以上。
RetCompletion:High-Speed Inference Image Completion with Retentive Network
- 采用双向序列融合与单向像素更新策略,兼顾结构一致性和生成效率。
- 在ImageNet和CelebA-HQ上推理速度比ICT快10倍、比RePaint快15倍。
- 适合需要实时图像修复的场景,如视频编辑或交互式设计。
时间开销是实现高质量多解图像补全的主要挑战。近期自然语言处理中的保留网络(Retentive Network, RetNet)因其低成本推理能力提供了新思路。受此启发,我们将RetNet引入计算机视觉的多解图像补全任务。提出RetCompletion,一种两阶段框架:第一阶段引入Bi-RetNet,通过双向序列信息融合整合图像上下文;推理时采用单向像素逐点更新策略,恢复一致图像结构,实现高重建质量与高速推理。第二阶段使用CNN进行低分辨率上采样以增强纹理细节。在ImageNet和CelebA-HQ上的实验表明,该方法推理速度比ICT快10倍,比RePaint快15倍。RetCompletion显著提升推理速度并保持优异性能。
原文摘要 · Abstract (English)
Time cost is a major challenge in achieving high-quality pluralistic image completion. Recently, the Retentive Network (RetNet) in natural language processing offers a novel approach to this problem with its low-cost inference capabilities. Inspired by this, we apply RetNet to the pluralistic image completion task in computer vision. We present RetCompletion, a two-stage framework. In the first stage, we introduce Bi-RetNet, a bidirectional sequence information fusion model that integrates contextual information from images. During inference, we employ a unidirectional pixel-wise update strategy to restore consistent image structures, achieving both high reconstruction quality and fast inference speed. In the second stage, we use a CNN for low-resolution upsampling to enhance texture details. Experiments on ImageNet and CelebA-HQ demonstrate that our inference speed is 10$\times$ faster than ICT and 15$\times$ faster than RePaint. The proposed RetCompletion significantly improves inference speed and delivers strong performance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。