Visual Prompt Tuning
This provides a more efficient method for adapting pre-trained vision models to downstream tasks, benefiting researchers and practitioners in computer vision.
The paper tackles the inefficiency of full fine-tuning for large vision Transformers by introducing Visual Prompt Tuning (VPT), which adds less than 1% trainable parameters and outperforms full fine-tuning in many cases while reducing storage costs.
The current modus operandi in adapting pre-trained models involves updating all the backbone parameters, ie, full fine-tuning. This paper introduces Visual Prompt Tuning (VPT) as an efficient and effective alternative to full fine-tuning for large-scale Transformer models in vision. Taking inspiration from recent advances in efficiently tuning large language models, VPT introduces only a small amount (less than 1% of model parameters) of trainable parameters in the input space while keeping the model backbone frozen. Via extensive experiments on a wide variety of downstream recognition tasks, we show that VPT achieves significant performance gains compared to other parameter efficient tuning protocols. Most importantly, VPT even outperforms full fine-tuning in many cases across model capacities and training data scales, while reducing per-task storage cost.