Apricity-cxr/Cascaded-Classification-Segmentation-Framework

stars

commits

Python

primary language

Aug 31, 2026

updated

README

Model Inference & Training Suite

Overview

This repository provides a modular and efficient inference pipeline for our pretrained models. Currently, we release inference code for both classification and segmentation tasks, allowing you to easily run predictions on your own images or videos.

What’s next?
We are actively working on releasing the full training code for both classification and segmentation models. Once open‑sourced, you will be able to fine‑tune or retrain the models entirely on your own custom datasets, with support for various backbones, data augmentations, and training strategies.

Contributors

Apricity-cxr

4 commits

Similar repos

juntaoJianggavin/APRIL-MedSeg

A Modern Modular 2D Medical Image Segmentation Toolbox

qubvel/segmentation_models.pytorch

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

11,732

lingxitong/PFM_Segmentation

Pathology Foundation Models Meet Semantic Segmentation

Vinicius-GN/segment_net

Flexible and modular framework for semantic segmentation using various transformer-based,…

autodistill/autodistill

Images to inference with no labeling (use foundation models to train supervised models).

2,772

ArgoHA/D-FINE-seg

D-FINE-seg Object Detection and Segmentation Framework (Train, Export, Inference)

203

xk-huang/segment-caption-anything

[CVPR'24] The repository provides code for running inference and training for "Segment and Caption…

232

med-air/Endo-FM

[MICCAI'23] Foundation Model for Endoscopy Video Analysis via Large-scale Self-supervised Pre-train

233

Languages

Python

100.0%

README

Model Inference & Training Suite

Overview

What’s next?
We are actively working on releasing the full training code for both classification and segmentation models. Once open‑sourced, you will be able to fine‑tune or retrain the models entirely on your own custom datasets, with support for various backbones, data augmentations, and training strategies.