StegFormer: Rebuilding the Glory of Autoencoder-Based Steganography

Authors: Xiao Ke, Huanqi Wu, Wenzhong Guo

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments demonstrate that our Steg Former outperforms existing state-of-the-art (SOTA) models.
Researcher Affiliation Academia 1Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing, College of Computer and Data Science, Fuzhou University, Fuzhou 350116, China 2Engineering Research Center of Big Data Intelligence, Ministry of Education, Fuzhou 350116, China kex@fzu.edu.cn, wuhuanqi135@gmail.com, guowenzhong@fzu.edu.cn
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes Our code will be released in https://github.com/aoli-gei/StegFormer.
Open Datasets Yes We use DIV2K to train our Steg Former and the testing datasets comprise DIV2K (Agustsson and Timofte 2017), COCO (Lin et al. 2014), and Image Net (Deng et al. 2009) to test the generalization ability.
Dataset Splits No The paper mentions using DIV2K for training and COCO/ImageNet for testing, but it does not specify explicit training/validation/test splits within these datasets or for the experiment setup.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments.
Software Dependencies No The paper mentions using 'The Adam W optimizer' but does not provide specific version numbers for any software dependencies or libraries.
Experiment Setup Yes The Adam W optimizer is used to train Steg Former with the cosine decay strategy to decrease the learning rate to 1e-6 with the initial learning rate 1e-3.