For　the　purpose　of　exploiting　the　intra-view　and　inter-view　dependencies　of　stereo　video,　a　fast　motion　and　disparity　estimation　is　proposed　in　this　paper.　The　key　idea　is　to　perform　iterative　search　in　both　the　motion　and　disparity　fields　based　on　the　stereo-motion　consistency　constraint　model,　and　adaptively　refine　the　window　size　according　to　the　model　errors　after　each　iteration　to　ensure　the　vector　accuracy.　In　this　way,　both　the　motion　and　disparity　vectors　can　be　obtained　quickly.　The　experimental　results　demonstrate　that　the　proposed　algorithm　can　reduce　coding　time　dramatically　at　the　cost　of　a　negligible　quality　decreasing.