There are techniques like passive radar, UWB and so on.

http://www.eurekalert.org/pub_releases/2009-10/uou-rw100909.php

http://www.newscientist.com/article/mg21128225.200-chaos-radar-uses-messy-signals-to-see-through-walls.html

http://web.mit.edu/newsoffice/2011/ll-seeing-through-walls-1018.html

http://researchnews.osu.edu/archive/noiserad.htm

http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?reload=true&arnumber=5613314&tag=1

http://scienceblog.com/10881/stealth-radar-sees-through-trees-walls-undetected/

http://www.eurekalert.org/pub_releases/2008-02/uosc-sbo020108.php

Regards,

Joachim

http://home.dei.polimi.it/savazzi/radio_imaging.htm

http://puma.isti.cnr.it/dfdownloadnew.php?ident=/cnr.isti/2012-TR-007&langver=en&scelta=NewMetadata

Thermal cameras (e.g. FLIR)

http://www.youtube.com/watch?v=1MNXZkoUKuM

http://lrm.icmc.usp.br/wiki/images/6/68/CBSEC2011_DSales.pdf

http://www.flir.com/cs/emea/en/view/?id=43139

And of course laser range sensors, right? You can take a look at one of my papers which is human detection from laser range data only. Of course, in general you can always extend methods by fusing not only sensors at a logic-algorithmic level but also at the representations (i.e. find a good unified space).