All Projects → namtpham → casia2groundtruth

namtpham / casia2groundtruth

Licence: other
Groundtruth images of tampering dataset CASIA 2.0

Projects that are alternatives of or similar to casia2groundtruth

casia1groundtruth
Groundtruth images of tampering dataset CASIA 1.0
Stars: ✭ 27 (-72.16%)
Mutual labels:  image-manipulation, casia, forgery, forgery-detection, groundtruth, splicing-detection, groundtruth-dataset
ManTraNet-pytorch
Implementation of the famous Image Manipulation\Forgery Detector "ManTraNet" in Pytorch
Stars: ✭ 47 (-51.55%)
Mutual labels:  image-manipulation, forgery, forgery-detection
Pel
PHP Exif Library - library for reading and writing Exif headers in JPEG and TIFF files using PHP.
Stars: ✭ 232 (+139.18%)
Mutual labels:  image-manipulation
dagpi
Dagpi is a powerful and fast api that does image manipulation as well as serves datasets. It is fast and written in rust and python. Perfect for discord bots, social media apps, camera apps and more.
Stars: ✭ 25 (-74.23%)
Mutual labels:  image-manipulation
idg
Document image generator
Stars: ✭ 40 (-58.76%)
Mutual labels:  image-manipulation
CoCosNet-v2
CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation
Stars: ✭ 312 (+221.65%)
Mutual labels:  image-manipulation
imtool
🖼️ Client-side canvas-based image manipulation library.
Stars: ✭ 38 (-60.82%)
Mutual labels:  image-manipulation
Pytorch Cyclegan And Pix2pix
Image-to-Image Translation in PyTorch
Stars: ✭ 16,477 (+16886.6%)
Mutual labels:  image-manipulation
Nova-Bot
This bot was created using discord.js v12. This bot offers you many powerful commands ranging from moderation to economy.
Stars: ✭ 37 (-61.86%)
Mutual labels:  image-manipulation
markerjs2
Add image annotation to your web apps.
Stars: ✭ 107 (+10.31%)
Mutual labels:  image-manipulation
rgb-editor
Tool to mass modify R, G ,B values for .png files, useful for making Minecraft Texturepacks
Stars: ✭ 15 (-84.54%)
Mutual labels:  image-manipulation
Awesome-low-level-vision-resources
A curated list of resources for Low-level Vision Tasks
Stars: ✭ 35 (-63.92%)
Mutual labels:  image-manipulation
js-image-carver
🌅 Content-aware image resizer and object remover based on Seam Carving algorithm
Stars: ✭ 1,467 (+1412.37%)
Mutual labels:  image-manipulation
imagekit-php
PHP SDK for ImageKit.io API.
Stars: ✭ 34 (-64.95%)
Mutual labels:  image-manipulation
Finegan
FineGAN: Unsupervised Hierarchical Disentanglement for Fine-grained Object Generation and Discovery
Stars: ✭ 240 (+147.42%)
Mutual labels:  image-manipulation
hermitage
Service that provides storage, delivery and modification of your images
Stars: ✭ 33 (-65.98%)
Mutual labels:  image-manipulation
Graphite
Open source 2D node-based raster/vector graphics editor (Photoshop + Illustrator + Houdini = Graphite)
Stars: ✭ 223 (+129.9%)
Mutual labels:  image-manipulation
libpillowfight
Small library containing various image processing algorithms (+ Python 3 bindings) that has almost no dependencies -- Moved to Gnome's Gitlab
Stars: ✭ 60 (-38.14%)
Mutual labels:  image-manipulation
Image-Processing-CLI-in-Rust
CLI for image processing with histograms, binary treshold and other functions
Stars: ✭ 25 (-74.23%)
Mutual labels:  image-manipulation
Fourier-and-Images
Fourier and Images
Stars: ✭ 81 (-16.49%)
Mutual labels:  image-manipulation

Groundtruth of spliced images in dataset CASIA 2.0

Tp_D_CRN_M_N_pla00035_pla00033_10997.jpg Tp_D_CRN_M_N_pla00035_pla00033_10997_gt.png

Tp_D_CRN_S_N_nat00033_cha00086_11502.jpg Tp_D_CRN_S_N_nat00033_cha00086_11502_gt.png

  • Groundtruth dataset can be downloaded directly from this repository.

  • Recently, I received several requests of the original dataset since the server is no longer available. I upload this dataset to my Drive to spread this dataset to the research community. Please visit the following link to download: (~2.6 GB)

  • [Update 2022/04/14] Revised dataset: https://bit.ly/2QazgkG

I removed the download link of the original dataset to avoid the confusion. If you need it, feel free to email me.

  • [Update 2019/09/30]: In this version, almost the noises in the previous version are handled. The file names are also revised carefully.

  • List of duplicate authentic images provided by another researcher.

Please notice that the authors made many mistakes in naming the files. People who download the original dataset (before 2021/04/20) please rename the tampered images using the commands in the excel files. Originally, it was reported that there are 5123 tampered images, including 3274 copy-move images and 1849 spliced images. However, mistakenly classified files are renamed as follows.

No. of images Originally named as Re-classified as
39 Copy-move images (TP_S_) Spliced images (S->D)
60 Spliced images (TP_D_) Copy-move images (D->S)
    After renaming the files, the number of copy-move and spliced images are 3295 and 1828, respectively.

Due to the lack of manual file, I write up here the naming convention:

Authentic images:

Au_ani_00001.jpg Au: Authentic ani: animal category Other categories: arc (architecture), art, cha (characters), ind (indoor), nat (nature), pla (plants), txt (texture)

Tampering images

a. Spliced image

    Tp_D_CRN_S_N_cha00063_art00014_11818.jpg
  • Tp: Tampering
  • D: Different (means the tampered region was copied from the different image)
  • Next 5 letters stand for the techniques they used to create the images. Unfortunately, I don't remember exactly.
  • cha00063: the source image
  • art00014: the target image
  • 11818: tampered image ID

b. Copy-move images

    Tp_S_NRN_M_N_pla00020_pla00020_10988.jpg
  • Tp: Tampering
  • S: Same (means the tampered region was copied from the same image)
  • And the rest is similar to case a.

If you use the groundtruth dataset for a scientific publication, please cite the following papers:

  • CASIA dataset

      @inproceedings{Dong2013,
      doi = {10.1109/chinasip.2013.6625374},
      url = {https://doi.org/10.1109/chinasip.2013.6625374},
      year = {2013},
      month = jul,
      publisher = {{IEEE}},
      author = {Jing Dong and Wei Wang and Tieniu Tan},
      title = {{CASIA} Image Tampering Detection Evaluation Database},
      booktitle = {2013 {IEEE} China Summit and International Conference on Signal and Information Processing}
      }
    
  • CASIA groundtruth dataset

     @article{pham2019hybrid,
     title={Hybrid Image-Retrieval Method for Image-Splicing Validation},
     author={Pham, Nam Thanh and Lee, Jong-Weon and Kwon, Goo-Rak and Park, Chun-Su},
     journal={Symmetry},
     volume={11},
     number={1},
     pages={83},
     year={2019},
     publisher={Multidisciplinary Digital Publishing Institute}
     }
    
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].