Digital holography offers appealing features for microscopic and 3D imaging applications. Thus, the growing interest of digital holographic data has led to the constitution of huge amounts of data. Therefore, in order to efficiently store and transmit this data, the design of hologram compression schemes becomes a crucial issue. In this paper, we propose to encode two sets of data, referred to as shifted distance information, by applying a non separable vector lifting scheme. The idea consists in exploiting the redundancies existing between the two sets and building a content-adaptive hologram compression scheme by resorting to a sparse optimization technique. Experimental results show that the proposed method outperforms the state-of-the-art ones.