Geometry　image　is　a　kind　of　mesh　representation　method　which　presents　a　three-dimensional　geometric　model　with　a　regular　structure.　Because　of　such　a　uniform　representation,　geometry　image　has　become　an　important　tool　to　effectively　deal　with　complex　three-dimensional　models.　Unfortunately,　details　of　the　input　mesh　are　missing　or　incorrectly　reconstructed　because　of　the　large　stretch　of　the　parametrized　vertices.　To　solve　this　problem,　we　propose　a　new　geometry　image　method　adaptively　by　subdividing　particular　regions　of　the　mesh　via　an　area　stretching　metric.　We　find　such　regions　with　larger　distortion　by　calculating　the　area　stretching　metric　for　each　triangle,　and　determine　the　region　to　be　subdivided.　After　dealing　with　the　mesh　locally,　we　can　obtain　a　more　accurate　parametrization,　which　will　produce　higher　quality　reconstructions.　Experiments　show　that,　our　method　can　produce　reconstruction　models　with　better　quality　and　achieves　better　PSNR　values.　1548-7741/Copyright　©　2014　Binary　Information　Press.