TRIST: Circumventing Censorship with Transcoding-Resistant Image Steganography

Chris Connolly explained that TRIST attempts to subvert statelevel censorship by encoding hidden messages steganographically in images and video. However, because images and video are posted to social networks (Facebook or YouTube), naively encoded hidden messages are often lost in the transcoding/compressing done by the social networks. TRIST attempts to solve this by using an encoding that survives this type of re-encoding. The encoding is even robust against resizing and filtering (e.g., sharpening filter). The bandwidth of this encoding is dependent on the highfrequency textures present in the cover image.