GLID: Pre-training a Generalist Encoder-Decoder Vision Model

arXiv:2404.07603v1 Announce Type: new
Abstract: This paper proposes a GeneraLIst encoder-Decoder (GLID) pre-training method for better handling various downstream computer vision tasks. While self-supervised pre-training approaches, e.g., Masked Autoencoder, have shown success in transfer learning, task-specific sub-architectures are still required to be appended for different downstream tasks, which cannot enjoy the benefits of large-scale pre-training. GLID overcomes this challenge by allowing the pre-trained generalist encoder-decoder to be fine-tuned on various vision tasks with minimal task-specific architecture modifications. In the GLID …

abstract architectures arxiv autoencoder benefits computer computer vision cs.cv decoder encoder encoder-decoder masked autoencoder paper pre-training success tasks training transfer transfer learning type vision