searching for words between delimeters from the rear

Hi,
i need to pick up dates and times from the file names which are of unequal length. The dates and time are delimited by dot. I am interested in getting the strings between the delimeter for fields -3, -4, -5 from behind (rear) so that the out put looks like :

071118.011300.556

I have tried with awk -F'.' '{print $3 '.'$4 '.'$5'}' filename
There is no consistency in the extractiion and i dont know how to read from the back which would have been more consistent.

See examples of filenames below:

nwg_0x005f_200611501-071118.011300.556.d3v4r5.bk

499c-9b4c-7808972f621d.7808972f621d.7808972f621d.7808972f621d.081222.201727.977.d3v70r764.bk

_cheshir.8c671a28-9a9d-42e9-96fd-02304c85bbf3.02304c85bbf3.02304c85bbf3.02304c85bbf3.090219.183923.709.d3v89r1353.bk

HC-071128.002729.081.d3v46r195.bk

al_0x005f_Publish-070724.195338.244.d3v9r20.bk

I need help.

Regards

Give in and out example. Easier to give solution.

sed 's/.*\([0-9]\{6\}.[0-9]\{6\}.[0-9]\{3\}\).*/\1/' file

Please use [code] tags when you post code or data sample.