From 44ce6f04372b024ef6afa8a275df1fd10b3fc2ca Mon Sep 17 00:00:00 2001 From: YadiraF Date: Fri, 23 Mar 2018 20:34:37 +0800 Subject: [PATCH] add pose --- README.md | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index fef4050..ca55e3b 100644 --- a/README.md +++ b/README.md @@ -14,7 +14,7 @@ The main features are: * **End-to-End** our method can directly regress the 3D facial structure and dense alignment from a single image bypassing 3DMM fitting. -* **Multi-task** By regressing position map, the 3D geometry along with semantic meaning can be obtained. Thus, we can effortlessly complete the tasks of dense alignment, monocular 3D face reconstruction, etc. +* **Multi-task** By regressing position map, the 3D geometry along with semantic meaning can be obtained. Thus, we can effortlessly complete the tasks of dense alignment, monocular 3D face reconstruction, pose estimation, etc. * **Faster than real-time** The method can run at over 100fps(with GTX 1080) to regress a position map. @@ -44,6 +44,10 @@ Get the 3D vertices and corresponding colours from a single image. Save the res * #### 3D Pose Estimation + Rather than only use 68 key points to calculate the camera matrix(easily effected by expression and poses), we use all vertices(more than 40K) to calculate a more accurate pose. + + #### ![pose](Docs/images/pose.jpg) + * #### Texture Fusion @@ -79,6 +83,7 @@ cd PRNet ```bash python run_basics.py #Can run only with python and tensorflow +python run.py # need dlib and opencv(show the alignment and pose results). ```