Skip to content

Latest commit

 

History

History
69 lines (58 loc) · 1.85 KB

README.md

File metadata and controls

69 lines (58 loc) · 1.85 KB

Towards Robust Image Stitching: An Adaptive Resistance Learning against Compatible Attacks

This repo is the official implementation of, “Towards Robust Image Stitching: An Adaptive Resistance Learning against Compatible Attacks”, Zhiying Jiang, Xingyuan Li, Jinyuan Liu, Xin Fan, Risheng Liu*, Association for the Advancement of Artificial Intelligence (AAAI), 2024.

Overview

avatar

Prerequisites

  • Linux or macOS
  • Python 3
  • CPU or NVIDIA GPU + CUDA CuDNN

🔑 Installation

Type the command:

pip install -r requirements.txt

🤖 Download

Download the pre-trained model

Download our collected dataset:

Note

  • For training, an NVIDIA GPU is strongly recommended for speed.
  • You can train and test the model using the following commands, or choose the pre-trained checkpoint that we provide.
  • For using custom data and pre-trained models, you need to replace the paths in the code with your own local paths.

ImageAlignment

  • First, you need to train a image alignment model:
cd ImageAlignment
python train.py
  • To test a trained alignment network:
python test.py

ImageReconstruction

  • Second, you need to train a image reconstruction model:
cd ../ImageReconstruction
python train.py
  • To test a trained reconstruction network:
python test.py

Then you can get the final robust stitching results.