# High-resolution processing

A 7-billion-parameter vision Transformer model trained on 8 billion language-unlabeled web images, achieving exceptional visual representation capabilities through self-supervised learning

This model uses an improved vnet architecture for image-to-image processing, supporting tasks such as adversarial, aesthetic, and quality enhancement

AIM-v2 is an image encoder model based on the timm library, with a 3B parameter scale, suitable for image feature extraction tasks.

AIM-v2 is an image encoder model based on the timm library, suitable for image feature extraction tasks.

CLIP model image encoder based on ResNet50 architecture with 64x width expansion, using Global Average Pooling (GAP) strategy

A ResNet50x16 variant model based on the CLIP framework, focused on image feature extraction

DepthPro is a monocular depth estimation model capable of predicting depth from a single image.

A self-supervised feature representation model based on ConvNeXt-V2, pre-trained using the Fully Convolutional Masked Autoencoder (FCMAE) framework, suitable for image classification and feature extraction tasks.