C

Convnextv2 Huge.fcmae

Developed by timm
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.
Downloads 52
Release Time : 1/5/2023

Model Overview

This model is a high-performance convolutional neural network specifically designed for image feature extraction and classification tasks. It is pre-trained via self-supervised learning and does not include a pre-trained head, making it suitable for fine-tuning or use as a feature extractor.

Model Features

Self-supervised pre-training
Pre-trained using the Fully Convolutional Masked Autoencoder (FCMAE) framework, eliminating the need for large amounts of labeled data.
High-performance architecture
Based on the ConvNeXt-V2 architecture, optimized for computational efficiency and feature extraction capabilities.
Flexible application
Can be used as a feature extractor or for fine-tuning, adaptable to various computer vision tasks.
Large-scale parameters
Boasts 657.5M parameters, providing powerful feature representation capabilities.

Model Capabilities

Image feature extraction
Image classification
Generating image embeddings

Use Cases

Computer vision
Image classification
Used for classifying images, supporting recognition of multiple categories.
Performs excellently on the ImageNet-1k dataset.
Feature extraction
Extracts high-level feature representations from images for downstream tasks.
Can generate high-quality image embeddings.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase