Update README.md

README.md

@@ -66,6 +66,10 @@ For completeness, this will also load model `./models/colorization_release_v2_no
 
 (ii) If you wish to do the color conversion inside of the network, use prototxt `./models/alexnet_deploy.prototxt`. The input should be BGR images, non-mean centered, in [0,255]. You will have to follow Caffe installation (described in step (2) in the previous section).
 
+### Appendix ###
+
+To run the "real vs fake" Amazon Mechanical Turk test (Table 1 of the paper), see [this repository]( https://github.com/phillipi/AMT_Real_vs_Fake). See line 1 of the **Usage** section. Corresponding paths are: [Ours (full)](http://colorization.eecs.berkeley.edu/imgs/classrebal_turk_imgs_438000/), [Ours (class, no rebal)](http://colorization.eecs.berkeley.edu/imgs/classnorebal_turk_imgs_446000/) [Ours (L2)](http://colorization.eecs.berkeley.edu/imgs/regress_turk_imgs_534000/), [Ours (L2, ft from class)](http://colorization.eecs.berkeley.edu/imgs/regressft_turk_imgs_312000/), [Ground Truth](http://colorization.eecs.berkeley.edu/imgs/gt_imgs_0/)
+
 ### Citation ###
 If you find this model useful for your resesarch, please use this [bibtex](http://richzhang.github.io/colorization/resources/bibtex_eccv2016_colorization.txt) to cite.