Cross Modal Steganography

Cross-modal steganography focuses on concealing data of one type (e.g., audio) within a different type of cover media (e.g., an image), making the hidden information undetectable. Recent research emphasizes using implicit neural representations (INRs) to achieve this, enabling flexible embedding across various modalities like images, audio, and video, and significantly increasing the capacity for hidden data. This approach offers advancements over traditional methods, particularly in handling diverse data types and achieving high embedding rates. The resulting techniques have implications for secure communication, data hiding, and copyright protection.