Semantic segmentation models with 500+ pretrained convolutional and transformer-based backbones.

PyTorch and TensorFlow/Keras image models with automatic weight conversions and equal API/implementations - Vision Transformer (ViT), ResNetV2, EfficientNetV2, NeRF, SegFormer, MixTransformer, (planned...) DeepLabV3+, ConvNeXtV2, YOLO, etc.

This repository demonstrates how to use TensorFlow based SegFormer model in 🤗 transformers package.

This repository provides an overview of Segformer, architecture encoder in particular. Some details of Segformer can be misleaded, thus makes a short description here to help understand the model. Meanwhile, the code (Keras/TensorFlow) is also provided for supporting.

This is a warehouse for SegFormer-pytorch-model, can be used to train your image datasets for segmentation tasks.

A Keras 3 Implementation of "SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers"

PyTorch Implementation of SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers

Real time Semantic Segmentation on Cityscapes dataset.

Comparing CNN- and Transformer-Based Deep Learning Models for Semantic Segmentation of Remote Sensing Images. Specifically looking at different configurations of U-Net and SegFormer applied to Potsdam, Vaihingen and FloodNet dataset.

Tackling the challenges of off-road environment navigation through Attention Guided Off-Road Semantic Segmentation.

Clothes semantic segmentation with Transformers

An application for automatic road damage assessment using semantic segmentation on high-resolution images. The project helps municipal authorities and maintenance teams detect and prioritize road repairs, improving safety and reducing costs.

Automatic Damage Segmentation of Sanskrit Palm-leaf Manuscripts

Cloud Cover Detection using Deep Learning

Semantic Segmentation of Drone Images With Hugging Face and MLflow

Repository of the master's thesis focused on the Automatic Detection of Objects in the Track

This project leverages the Segformer pretrained model to perform billboard segmentation through semantic segmentation, specifically targeting Indian billboards. By applying transfer learning on a custom dataset, the model accurately classifies and delineates billboards within images, enabling efficient detection and analysis.

A machine learning project designed to detect road damage from images, leveraging deep learning and computer vision techniques for efficient and accurate damage detection.

Breast Cancer Ultrasound Segmentation using Deep Learning

A repository of SegFormer model test