In　the　past　few　years,　data　traffic　generated　by　video　applications　has　become　the　dominant　traffic　on　the　internet　due　to　the　rapid　development　of　wireless　network　and　mobile　terminal.　However,　using　a　single　wireless　network　for　video　transmission　with　limited　bandwidth　is　facing　great　challenges　under　the　increasing　video　quality　requirement.　Fortunately,　mobile　terminal　equipped　with　multiple　network　interfaces　can　transmit　data　over　heterogeneous　wireless　networks　concurrently　to　achieve　higher　bandwidth.　In　order　to　make　effective　use　of　heterogeneous　access　network　resources,　this　paper　develops　a　heuristic　flow　splitting　algorithm　based　on　ACO　(Ant　Colony　Optimization)　algorithm.　Firstly,　a　system　model　with　the　objective　that　minimizes　the　multipath　video　transmission　distortion　is　proposed　under　the　end-to-end　delay　constraint.　Secondly,　to　solve　the　problem　based　on　heuristic　ACO,　the　relevant　models　such　as　pheromone　model　and　reliability　model　are　presented,　and　a　heuristic　ACO　algorithm　is　proposed.　Finally,　the　proposed　algorithm　is　tested　and　evaluated　in　the　real-world　platform.　The　evaluation　results　show　that　the　proposed　algorithm　performs　better　than　the　referenced　traditional　algorithm　in　terms　of　PSNR　and　end-to-end　frame　delay.