Skip to content

Latest commit

 

History

History
28 lines (21 loc) · 1.85 KB

README.md

File metadata and controls

28 lines (21 loc) · 1.85 KB
Raw

Parallel Computing Accelerated Image Inpainting using GPU CUDA, Theano, and Tensorflow

By Heronimus Tresy Renata Adie, Ignatius Aldi Pradana, Pranowo (UAJY)

This work was published in 2018 10th International Conference on Information Technology and Electrical Engineering (ICITEE). The official full paper available on ResearchGate (DOI: 10.1109/ICITEED.2018.8534858 ).

*All the completes codes, datasets, and step to reproduce will be published here soon.

Introduction

PDE Image Inpainting method running on GPU with CUDA, Tensorflow, and Theano. This work show benchmark performance of PDE image inpainting running on CPU using C++, Theano, and Tensorflow and on GPU with CUDA, Theano, and Tensorflow. The benchmark shows that parallel computing accelerated PDE image inpainting can run faster on GPU either with CUDA, Theano, or Tensorflow compared to PDE image inpainting running on CPU.

License

Parallel Computing Accelerated Image Inpainting using GPU CUDA, Theano, and Tensorflow is released under the MIT license.

Citing

If you find this work useful in your research, please consider citing:

@inproceedings{inproceedings,
    author = {Tresy Renata Adie, Heronimus and Aldi Pradana, Ignatius and Pranowo, Pranowo},
    year = {2018},
    month = {07},
    pages = {621-625},
    title = {Parallel Computing Accelerated Image Inpainting using GPU CUDA, Theano, and Tensorflow},
    doi = {10.1109/ICITEED.2018.8534858}
}

Author Contact

Heronimus Tresy Renata Adie : [email protected], LinkedIn, about.me/heronimus.