In　this　paper,　a　new　multiple　description　image　coding　scheme　using　directional　lifting　transform　is　proposed.　The　basic　idea　is　to　divide　an　image　into　two　descriptions　with　quincunx　segmentation.　The　traditional　spatial　domain　multiple　description　image　coding　techniques　usually　result　in　the　very　low　coding　efficiency.　To　tackle　this　problem,　we　propose　a　directional　lifting　technique　to　extensively　exploit　the　correlations　among　neighboring　pixels.　The　lifting　directions　are　selected　and　organized　based　on　the　quadtree　partition.　In　addition,　a　novel　interpolation　scheme　is　employed　in　the　side　decoder　to　reconstruct　one　description　with　full　resolution.　As　for　the　central　decoder　operated　on　the　two　descriptions,　data　fusion　algorithm　is　employed　to　reconstruct　the　whole　image.　Thus,　the　visual　quality　from　the　central　decoder　can　be　further　improved.　©　Springer-Verlag　Berlin　Heidelberg　2005.