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

TL;DR

EfficientViT introduces a multi-scale linear attention mechanism for high-resolution dense prediction, achieving global receptive fields and multi-scale learning with lightweight operations, resulting in significant speedups across various hardware platforms.

Contribution

The paper proposes a novel multi-scale linear attention method for high-resolution vision models, improving efficiency and performance over existing methods without sacrificing accuracy.

Findings

Up to 13.9x GPU latency reduction on Cityscapes

6.4x speedup in super-resolution tasks

48.9x higher throughput on A100 GPU

Abstract

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.