title update

LICENSE

@@ -40,7 +40,7 @@ You acknowledge that the Data & Software is a valuable scientific resource and a
 Citation:
 
 @inproceedings{xiu2023econ,
-  title     = {{ECON: Explicit Clothed humans Obtained from Normals}},
+  title     = {{ECON: Explicit Clothed humans Optimized via Normal integration}},
   author    = {Xiu, Yuliang and Yang, Jinlong and Cao, Xu and Tzionas, Dimitrios and Black, Michael J.},
   booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
   month     = {June},

README.md

@@ -2,7 +2,7 @@
 
 <p align="center">
 
-  <h1 align="center">ECON: Explicit Clothed humans Obtained from Normals</h1>
+  <h1 align="center">ECON: Explicit Clothed humans Optimized via Normal integration</h1>
   <p align="center">
     <a href="http://xiuyuliang.cn/"><strong>Yuliang Xiu</strong></a>
     ·
@@ -152,7 +152,7 @@ python -m apps.avatarizer -n <filename>
 
 ```bibtex
 @inproceedings{xiu2023econ,
-  title     = {{ECON: Explicit Clothed humans Obtained from Normals}},
+  title     = {{ECON: Explicit Clothed humans Optimized via Normal integration}},
   author    = {Xiu, Yuliang and Yang, Jinlong and Cao, Xu and Tzionas, Dimitrios and Black, Michael J.},
   booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
   month     = {June},

environment.yaml

@@ -9,7 +9,7 @@ channels:
   - defaults
 dependencies:
   - python=3.8
-  - pytorch-cuda=11.7
+  - pytorch-cuda=11.6
   - pytorch=1.13.0
   - nvidiacub
   - torchvision

lib/common/render.py

@@ -315,7 +315,7 @@ class Render:
             save_path,
             fourcc,
             self.fps,
-            (width, int(height * 1.5)),
+            (width*3, int(height)),
         )
 
         pbar = tqdm(range(len(self.meshes)))
@@ -358,15 +358,9 @@ class Render:
             img_cloth = blend_rgb_norm(
                 (torch.stack(mesh_renders)[num_obj:, cam_id] - 0.5) * 2.0, data
             )
-
-            top_img = cv2.resize(
-                torch.cat([img_raw, img_smpl],
-                          dim=-1).squeeze(0).permute(1, 2, 0).numpy().astype(np.uint8),
-                (width, height // 2)
-            )
-            final_img = np.concatenate(
-                [top_img, img_cloth.squeeze(0).permute(1, 2, 0).numpy().astype(np.uint8)], axis=0
-            )
+            final_img = torch.cat(
+                [img_raw, img_smpl, img_cloth], dim=-1).squeeze(0).permute(1, 2, 0).numpy().astype(np.uint8)
+            
             video.write(final_img[:, :, ::-1])
 
         video.release()

lib/dataset/mesh_util.py

@@ -388,7 +388,7 @@ def poisson(mesh, obj_path, depth=10, decimation=True):
     pcl = o3d.io.read_point_cloud(pcd_path)
     with o3d.utility.VerbosityContextManager(o3d.utility.VerbosityLevel.Error) as cm:
         mesh, densities = o3d.geometry.TriangleMesh.create_from_point_cloud_poisson(
-            pcl, depth=depth, n_threads=6
+            pcl, depth=depth, n_threads=-1
         )
 
     # only keep the largest component