Mercurial > thoughts
diff org/adelson-notes.org @ 89:09d093d9b61f
merge ideas together.
author | Robert McIntyre <rlm@mit.edu> |
---|---|
date | Mon, 10 Mar 2014 13:01:48 -0400 |
parents | dfcbbb3d4b9a |
children |
line wrap: on
line diff
1.1 --- a/org/adelson-notes.org Mon Mar 10 12:59:38 2014 -0400 1.2 +++ b/org/adelson-notes.org Mon Mar 10 13:01:48 2014 -0400 1.3 @@ -66,8 +66,6 @@ 1.4 1.5 Horn, technical report, 1970 1.6 1.7 - 1.8 - 1.9 * Fri Oct 4 2013 1.10 1.11 Student report. Talked about how you capture the appearance of a 1.12 @@ -93,6 +91,26 @@ 1.13 objects made of different materials. The images were gathered off of 1.14 Flikr. 1.15 1.16 - 1.17 + Then she gave another talk from someone else. It's about assigning 1.18 + materials to objects and then rendering them. The choice of 1.19 + materials is determined by some sort of expert system? 1.20 1.21 - 1.22 + They have made a neat looking interface for human entry of texture 1.23 + labeling or objects in scenes. The important elements were manual 1.24 + labels, dynamic display of the current selection, and undo. 1.25 + 1.26 + There are papers about mechanical turk engineering. 1.27 + 1.28 + lol http://opensurfaces.cs.cornell.edu/ 1.29 + 1.30 + CUBAM is interesting 2010 "The multidimensional wisdom of crowds." 1.31 + "Neural Information processing systems" some sort of voting scheme. 1.32 + 1.33 + The point of this is apparently to do some kitchen makeover 1.34 + thing. You would take a picture of your kitchen, or you would look 1.35 + for a kitchen that looks like yours, and then you would be able to 1.36 + investigate different textures for your own kitchen. 1.37 + 1.38 + apparently "Label ME" has never been appropriately 1.39 + crowdsourced. Turns out that you get better work if you don't use 1.40 + idiots to do the work lol. 1.41 \ No newline at end of file