updated to reflect new pretrained model

README.md

@@ -3,6 +3,8 @@
 
 This project applies an implementation of [Globally and Locally Consistent Image Completion](http://hi.cs.waseda.ac.jp/%7Eiizuka/projects/completion/data/completion_sig2017.pdf) to the problem of hentai decensorship. Using a deep fully convolutional neural network, DeepMindBreak can replace censored artwork in hentai with plausible reconstructions. The user needs to only specify the censored regions.
 
+March 14th, 2018: Updated pretrained model to handle censor bars in any orientation.
+
 # **THIS PROJECT IS STILL IN DEVELOPMENT. DO NOT BE DISAPPOINTED IF THE RESULTS AREN'T AS GOOD AS YOU EXPECT.**
 
 ![Censored, decensored](/readme_images/collage.png)
@@ -25,8 +27,6 @@ It does NOT work with:
 
 In particular, if a vagina or penis is completely censored out, inpainting will be ineffective.
 
-Embarrassingly, because the neural network was trained to decensor horizontally and vertically oriented rectangles, it has trouble with angled rectangles. This will be fixed soon.
-
 # Dependencies
 
 - Python 2/3
@@ -34,7 +34,7 @@ Embarrassingly, because the neural network was trained to decensor horizontally
 - Pillow
 - tqdm
 - scipy
-- pyamg (only needed if poisson blending is enabled in decensor.py)
+- pyamg
 - matplotlib (only for running test.py)
 
 No GPU required! Tested on Ubuntu 16.04 and Windows.