Pyramid Vision Transformer (PVT) model pre-trained on ImageNet-1K (1 million images, 1000 classes) at resolution 224x224, and fine-tuned on ImageNet 2012 (1 million images, 1,000 classes) at resolution 224x224. It was introduced in the paper Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions by Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao and first released in this repository.

Disclaimer: The team releasing PVT did not write a model card for this model so this model card has been written by Rinat S. [@Xrenya].

The Pyramid Vision Transformer (PVT) is a transformer encoder model (BERT-like) pretrained on ImageNet-1k (also referred to as ILSVRC2012), a dataset comprising 1 million images and 1,000 classes, also at resolution 224x224.

Images are presented to the model as a sequence of variable-size patches, which are linearly embedded. Unlike ViT models, PVT is using a progressive shrinking pyramid to reduce computations of large feature maps at each stage. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.

By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.

You can use the raw model for image classification. See the model hub to look for fine-tuned versions on a task that interests you.

Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:

For more code examples, we refer to the documentation.

The ViT model was pretrained on ImageNet-1k, a dataset consisting of 1 million images and 1k classes.

The exact details of preprocessing of images during training/validation can be found here.

Images are resized/rescaled to the same resolution (224x224) and normalized across the RGB channels with mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225).