RGB-encoded normals	Needle map	Albedo map	Reconstructed surface without albedos 1	Reconstructed surface without albedos 2	Reconstructed surface with albedos 1	Reconstructed surface with albedos 2

Apart from very slight differences between the my program and the example program with respect to the light source vectors, my program produces identical results. Filling in the functions was straightforward given the hints and the email discussion which offered a way to deal with "undefined" normals. The method could be improved in a number of ways. To name two: a more realistic lighting model perhaps directly incorporating some specular reflection component would could improve the results, a more appropriate weighting than the image intensity for computing normals could yield better results.