Minor refactor

Author: radekd91

README.md

@@ -25,7 +25,7 @@ Current infernal projects:
 - [TalkingHead](inferno_apps/TalkingHead) 
   - official release of [EMOTE: Emotional Speech- Driven Animation with C ontent- Emotion Disentanglement](https://emote.is.tue.mpg.de/index.html)
   - tools to run, train or finetune speech-driven 3D avatars 
-- [FaceReconstruction](inferno_apps/EMOCA)
+- [FaceReconstruction](inferno_apps/FaceReconstruction)
   - contains EMICA - a combination of [DECA](https://deca.is.tue.mpg.de/), [EMOCA](https://emoca.is.tue.mpg.de/), [SPECTRE](https://filby89.github.io/spectre/) and [MICA](https://zielon.github.io/mica/) which produces excellent results
   - tools to run, train or fine-tune state-of-the-art in-the-wild face reconstruction 
 - [MotionPrior](inferno_apps/MotionPrior) 

inferno_apps/TalkingHead/data_processing/download_processed_mead.sh

@@ -74,8 +74,8 @@ echo "Unzipping the detected ladmarks..."
 unzip -q landmarks.zip
 echo "Unzipping the recognized emotions..."
 unzip -q emotions.zip
-# echo "Unzipping the reconstructions used in the EMOTE paper"
-# unzip -q reconstruction_v0.zip
+echo "Unzipping the reconstructions used in the EMOTE paper"
+unzip -q reconstruction_v0.zip
 echo "Unzipping the newer version of the reconstructions..."
 unzip -q reconstruction_v1.zip
 

inferno_apps/TalkingHead/utils/load.py

@@ -37,11 +37,11 @@ def load_model(path_to_models,
         cfg.learning.losses = {}
         cfg.learning.metrics = {}
 
-    faceformer = load_faceformer(cfg, mode,)
-    return faceformer, cfg
+    talkinghead_model = load_talking_head(cfg, mode,)
+    return talkinghead_model, cfg
 
 
-def load_faceformer(
+def load_talking_head(
               cfg,
               mode,
               relative_to_path=None,