Underwater Networks poses as a challenging environment for video transmissions mainly because of an extremely limited available bitrate. In such constrained conditions, it is crucial to know the opinion of users which will settle the difference between an effective and a useless service. However, the hostilities of the underwater channel extend to Video Quality Assessment adding challenges specific to this service context. The present work explores the major issues in the standard approaches to quality assessment: subjective experiments where human users evaluate the service and objective mathematical models for opinion scores. The main problems are presented and some key directions for future research are identified.