Snap & NEU’s EfficientFormer Models Push ViTs to MobileNet Speeds While Maintaining High Performance

In the new paper EfficientFormer: Vision Transformers at MobileNet, a research team from Snap Inc. and Northeastern University proposes EfficientFormer, a vision transformer that runs as fast as MobileNet while maintaining high performance.

ai artificial intelligence computer vision & graphics deep-neural-networks machine learning machine learning & data science ml mobilenet performance research technology vision-transformer