Noise Doesn't Lie: Towards Universal Detection of Deep Inpainting
Authors: Ang Li, Qiuhong Ke, Xingjun Ma, Haiqin Weng, Zhiyuan Zong, Feng Xue, Rui Zhang
IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically show, on multiple benchmark datasets, that our approach outperforms existing detection methods by a large margin and generalize well to unseen deep inpainting techniques. |
| Researcher Affiliation | Collaboration | Ang Li1 , Qiuhong Ke1 , Xingjun Ma2 , Haiqin Weng3 , Zhiyuan Zong3 , Feng Xue3 and Rui Zhang4 1The University of Melbourne 2Deakin University 3Ant Group 4Tsinghua University |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement or link for open-source code availability for the described methodology. |
| Open Datasets | Yes | We use three different deep inpainting techniques including GL [Iizuka et al., 2017], CA [Yu et al., 2018] and GC [Yu et al., 2019] to generate inpainted images on two datasets Places2 [Zhou et al., 2017] and Celeb A [Liu et al., 2015]. |
| Dataset Splits | Yes | For each of the two datasets, we randomly select (without replacement) 50K, 10K and 10K images to create the training, validation and testing subsets respectively, following either our universal data generation or using one of the above three inpainting techniques (GL, CA and GC). |
| Hardware Specification | Yes | All of our experiments were run with a Nvidia Tesla V100 GPU. |
| Software Dependencies | No | The paper mentions optimizers and loss functions but does not specify software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow, CUDA versions). |
| Experiment Setup | Yes | We train the networks using the Adam optimizer with initial learning rate 1e-4. An early stopping strategy is also adopted based on the mIoU on the validation dataset: the model with the highest validation mIoU is saved as the final model. The focal loss Lf is defined as following: Lf = Pn i ( Mi(1 ˆ Mi)γ log ˆ Mi (1 Mi) ˆ M γ i log(1 ˆ Mi)) where γ is the focusing parameter and is set to 2. Finally, the detected mask ˆ Mi can be obtained by binarizing Pi according to a threshold value. In this paper, we set the threshold value as 0.5. |