# SparseViT2025

**Repository Path**: zxnvszyk/SparseViT2025

## Basic Information

- **Project Name**: SparseViT2025
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-02-27
- **Last Updated**: 2025-02-27

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Can We Get Rid of Handcrafted Feature Extractors? SparseViT: Nonsemantics-Centered, Parameter-Efficient Image Manipulation Localization through Spare-Coding Transformer

Official repository for the AAAI2025 paper  *Can We Get Rid of Handcrafted Feature Extractors? SparseViT: Nonsemantics-Centered, Parameter-Efficient Image Manipulation Localization through Spare-Coding Transformer* [[paper]](https://arxiv.org/abs/2412.14598) [[website]](https://github.com/scu-zjz/SparseViT).

<p float="left">
  <img src="images/sparsevit.png" alt="SparseViT Framework Diagram" width="98%">
</p>

In summary, SparseViT leverages the distinction between semantic and non-semantic features, enabling the model to adaptively extract non-semantic features that are more critical for image manipulation localization. This provides a novel approach to precisely identifying manipulated regions.

## Test setup (Code + Models)

<details>
  <summary>1) Set up the coding environment</summary>
  <br>

   - First, clone the repository:

  ```bash
git clone https://github.com/scu-zjz/SparseViT.git
  ```

  - Our environment

  ```
Ubuntu LTS 20.04.1

CUDA 11.5 + cudnn 8.4.0

Python 3.10

PyTorch 2.4
  ```

  - You should install the packages in [requirements.txt](https://github.com/scu-zjz/SparseViT/blob/main/requirements.txt) 

  ```bash
pip install -r requirements.txt
  ```

</details>

<details>
  <summary>2) Download our pretrained checkpoints</summary>
  <br>
  
  - Download our pretrained checkpoints from [Google Drive](https://drive.google.com/drive/folders/1v-8I1WCbR0hpaV434yPgsFiimu6wWLCW?usp=drive_link) and place them in the checkpoint directory.
    
</details>

## Scripts

This should be super easy! Simply run

```
python main_test.py
```

Here, we have simply provided a basic test of SparseViT. Of course, you can train and test SparseViT within our proposed [IMDL-BenCo](https://github.com/scu-zjz/IMDLBenCo) framework, as they are fully compatible.

## Citation

If you find our code useful, please consider citing us and give us a star! 

```
@misc{su2024can,
  title={Can We Get Rid of Handcrafted Feature Extractors? SparseViT: Nonsemantics-Centered, Parameter-Efficient Image Manipulation Localization Through Spare-Coding Transformer},
  author={Su, Lei and Ma, Xiaochen and Zhu, Xuekang and Niu, Chaoqun and Lei, Zeyu and Zhou, Ji-Zhe},
  year={2024},
  eprint={2412.14598},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}
```