arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2205.14756cs.CV
收藏

EfficientViT:用于高分辨率密集预测的多尺度线性注意力

EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction

MIT(麻省理工学院);Zhejiang University(浙江大学);Tsinghua University(清华大学);MIT-IBM Watson AI Lab(麻省理工学院-IBM沃森人工智能实验室)

Han Cai, Junyan Li, Muyan Hu, Chuang Gan, Song Han

AI总结:

EfficientViT通过多尺度线性注意力实现全局感受野与多尺度学习,以轻量高效操作在高分辨率密集预测中大幅加速并保持性能,适用于多种硬件平台。

摘要:

高分辨率密集预测支持许多引人注目的实际应用,例如计算摄影、自动驾驶等。然而,巨大的计算成本使得在硬件设备上部署最先进的高分辨率密集预测模型变得困难。本工作提出了EfficientViT,这是一个具有新颖多尺度线性注意力的高分辨率视觉模型新系列。与先前依赖重型softmax注意力、硬件效率低下的大核卷积或复杂拓扑结构以获得良好性能的高分辨率密集预测模型不同,我们的多尺度线性注意力仅通过轻量级且硬件高效的操作即可实现全局感受野和多尺度学习(高分辨率密集预测的两个理想特性)。因此,EfficientViT在多种硬件平台(包括移动CPU、边缘GPU和云GPU)上相较于先前最先进的模型取得了显著的性能提升,并实现了大幅加速。在Cityscapes上无性能损失的情况下,我们的EfficientViT相较于SegFormer和SegNeXt分别提供了高达13.9倍和6.2倍的GPU延迟降低。对于超分辨率,EfficientViT相较于Restormer提供了高达6.4倍的加速,同时在PSNR上获得了0.11dB的提升。对于Segment Anything,EfficientViT在A100 GPU上提供了48.9倍更高的吞吐量,同时在COCO上实现了略优的零样本实例分割性能。

英文摘要:

High-resolution dense prediction enables many appealing real-world applications, such as computational photography, autonomous driving, etc. However, the vast computational cost makes deploying state-of-the-art high-resolution dense prediction models on hardware devices difficult. This work presents EfficientViT, a new family of high-resolution vision models with novel multi-scale linear attention. Unlike prior high-resolution dense prediction models that rely on heavy softmax attention, hardware-inefficient large-kernel convolution, or complicated topology structure to obtain good performances, our multi-scale linear attention achieves the global receptive field and multi-scale learning (two desirable features for high-resolution dense prediction) with only lightweight and hardware-efficient operations. As such, EfficientViT delivers remarkable performance gains over previous state-of-the-art models with significant speedup on diverse hardware platforms, including mobile CPU, edge GPU, and cloud GPU. Without performance loss on Cityscapes, our EfficientViT provides up to 13.9$\times$ and 6.2$\times$ GPU latency reduction over SegFormer and SegNeXt, respectively. For super-resolution, EfficientViT delivers up to 6.4x speedup over Restormer while providing 0.11dB gain in PSNR. For Segment Anything, EfficientViT delivers 48.9x higher throughput on A100 GPU while achieving slightly better zero-shot instance segmentation performance on COCO.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑