These are historical "snapshots" of the results table. See the latest results here.
|2017_01_07||before rounding bug fix||We fixed a rounding issue on 1/8/2017. This affected the error numbers for the three "old" image pairs with integer ground truth (test/Computer, training/ArtL, and training/Teddy), for which submitted disparities should have been rounded to integers (at full resolution) before evaluation.|
|2015_09_14||ELAS update||Snapshot before updating ELAS results. The old ELAS results were worse because of a bug in the conversion to grayscale images (using ImageMagick's "convert" under linux). The conversion is now done via the code in the SDK 1.6.|
|2015_03_26||Evaluation code update||This snapshot was created before we updated the evaluation code and metrics.|