From a9e4271496ee8cd3e1d98c7dd0c3136dfe49a7e8 Mon Sep 17 00:00:00 2001 From: rdanecek <danekradek@gmail.com> Date: Mon, 13 Feb 2023 16:47:58 +0100 Subject: [PATCH] Fix typo --- gdl_apps/EMOCA/README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/gdl_apps/EMOCA/README.md b/gdl_apps/EMOCA/README.md index 4e9020b9..dc4176c3 100644 --- a/gdl_apps/EMOCA/README.md +++ b/gdl_apps/EMOCA/README.md @@ -31,8 +31,8 @@ There are now mutiple EMOCA models that you can use for reconstruction. You can The available models are: 1) `EMOCA` - the original `EMOCA version -2) `EMOCA_v2_mp` - EMOCA v2 trained with mediapiple landmarks (instead of FAN) but without the lip reading loss -3) `EMOCA_v2_lr_cos_1.5` - EMOCA v2 trained with mediapiple landmarks and with the lip reading loss (cosine similarity on lip reading features, similarly to SPECTRE) +2) `EMOCA_v2_mp` - EMOCA v2 trained with mediapipe landmarks (instead of FAN) but without the lip reading loss +3) `EMOCA_v2_lr_cos_1.5` - EMOCA v2 trained with mediapipe landmarks and with the lip reading loss (cosine similarity on lip reading features, similarly to SPECTRE) 4) `EMOCA_v2_lr_mse_20` - (default) EMOCA v2 trained with mediapipe landmarks and with the lip reading loss (MSE on lip reading features) Notes: -- GitLab