Existing generative image steganography methods typically encode secret information into latent vectors, which are transformed into the entangled features of generated images. This approach faces two main challenges: (1) Transmission can degrade the quality of stego-images, causing bit errors in information extraction. (2) High embedding capacity often reduces the accuracy of information extraction. To overcome these limitations, this paper presents a novel generative image steganography via encoding pose keypoints. This method employs an LSTM-based sequence generation model to embed secret information into the generation process of pose keypoint sequences. Each generated sequence is drawn as a keypoint connectivity graph, which serves as input with an original image to a trained pose-guided person image generation model (DPTN-TA) to generate an image with the target pose. The sender uploads the generated images to a public channel to transmit the secret information. On the receiver’s side, an improved YOLOv8 pose estimation model extracts the pose keypoints from the stego-images and decodes the embedded secret information using the sequence generation model. Extensive experiments on the DeepFashion dataset show that the proposed method significantly outperforms state-of-the-art methods in information extraction accuracy, achieving 99.94%. It also achieves an average hiding capacity of 178.4 bits per image. This method is robust against common image attacks, such as salt and pepper noise, median filtering, compression, and screenshots, with an average bit error rate of less than 0.87%. Additionally, the method is optimized for fast inference and lightweight deployment, enhancing its real-world applicability.

Steganography is a technique for hiding secret information in imperceptible carriers and transmitting it. Unlike traditional embedding-based steganography, generative steganography can generate stego-carriers directly from secret messages, thus avoiding modifications to natural carriers that steganalysis can detect. As a branch of generative steganography, game-behavior-based steganography transmits secret information by encoding game behavior. It can naturally integrate with real interaction scenarios, exhibiting strong concealment and undetectability. To this end, this paper proposes a generative steganography based on Chinese Chess record construction. Firstly, an AlphaZero model was trained to achieve a high level in Chinese Chess, then transmit secret information by encoding chess behavior. Specifically, in each chess step, the model generates all the current feasible moves and encodes the moves that meet the threshold strategy according to probability. Then, the appropriate move will be selected according to the secret information. To ensure the reasonableness of the generated chess records, this paper controlled the game process and designed a database of fixed opening chess records. The proposed method can hide an average of 413 bits of information for each carrier and effectively resist common image attacks. Regarding anti-steganalysis, the proposed method achieved accuracy rates of 0.498 and 0.497 on XuNet and YeNet, respectively, outperforming other behavior-based steganography techniques.