GUI and cache_clean for locals

app.py

@@ -26,33 +26,53 @@ from apps.infer import generate_model
 # running
 
 description = '''
-# ICON Clothed Human Digitization
+# ICON Clothed Human Digitization 
 ### ICON: Implicit Clothed humans Obtained from Normals (CVPR 2022)
 
-<table style="width:26%; padding:0; margin:0;">
-<tr>
-<th><iframe src="https://ghbtns.com/github-btn.html?user=yuliangxiu&repo=ICON&type=star&count=true&v=2&size=small" frameborder="0" scrolling="0" width="100" height="20"></iframe></th>
-<th><a href="https://twitter.com/yuliangxiu"><img alt="Twitter Follow" src="https://img.shields.io/twitter/follow/yuliangxiu?style=social"></a></th>
-<th><a href="https://youtu.be/hZd6AYin2DE"><img alt="YouTube Video Views" src="https://img.shields.io/youtube/views/hZd6AYin2DE?style=social"></a></th>
-</tr>
+<table>
+<th>
+<ul>
+<li><strong>Homepage</strong> <a href="http://icon.is.tue.mpg.de">icon.is.tue.mpg.de</a></li>
+<li><strong>Code</strong> <a href="https://github.com/YuliangXiu/ICON">YuliangXiu/ICON</a>
+<li><strong>Paper</strong> <a href="https://arxiv.org/abs/2112.09127">arXiv</a>, <a href="https://readpaper.com/paper/4569785684533977089">ReadPaper</a>
+<li><strong>Chatroom</strong> <a href="https://discord.gg/Vqa7KBGRyk">Discord</a>
+</ul>
+<a href="https://twitter.com/yuliangxiu"><img alt="Twitter Follow" src="https://img.shields.io/twitter/follow/yuliangxiu?style=social"></a>
+<iframe src="https://ghbtns.com/github-btn.html?user=yuliangxiu&repo=ICON&type=star&count=true&v=2&size=small" frameborder="0" scrolling="0" width="100" height="20"></iframe>
+<a href="https://youtu.be/hZd6AYin2DE"><img alt="YouTube Video Views" src="https://img.shields.io/youtube/views/hZd6AYin2DE?style=social"></a>
+</th>
+<th>
+<iframe width="560" height="315" src="https://www.youtube.com/embed/hZd6AYin2DE" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
+</th>
 </table>
 
+#### The reconstruction + refinement + video take about 80~120 seconds for single image.
+
+<details>
+
+<summary>More</summary>
+
+#### Citation
+```
+@inproceedings{xiu2022icon,
+  title     = {{ICON}: {I}mplicit {C}lothed humans {O}btained from {N}ormals},
+  author    = {Xiu, Yuliang and Yang, Jinlong and Tzionas, Dimitrios and Black, Michael J.},
+  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
+  month     = {June},
+  year      = {2022},
+  pages     = {13296-13306}
+} 
+```
+
 #### Acknowledgments:
 
 - [StyleGAN-Human, ECCV 2022](https://stylegan-human.github.io/)
 - [nagolinc/styleGanHuman_and_PIFu](https://huggingface.co/spaces/nagolinc/styleGanHuman_and_PIFu)
 - [radames/PIFu-Clothed-Human-Digitization](https://huggingface.co/spaces/radames/PIFu-Clothed-Human-Digitization)
 
-#### The reconstruction + refinement + video take about 80 seconds for single image.
-
-<details>
-
-<summary>More</summary>
-
 #### Image Credits
 
 * [Pinterest](https://www.pinterest.com/search/pins/?q=parkour&rs=sitelinks_searchbox)
-* [Qianli Ma](https://qianlim.github.io/)
 
 #### Related works
 
@@ -72,10 +92,9 @@ def generate_image(seed, psi):
     return img
 
 
-random.seed(1993)
+random.seed(2022)
 model_types = ['icon-filter', 'pifu', 'pamir']
-examples = [[item, random.choice(model_types)] for item in random.sample(
-    sorted(glob.glob('examples/*.png')), 8)]
+examples = [[item, random.choice(model_types)] for item in glob.glob('examples/*.png')]
 
 with gr.Blocks() as demo:
     gr.Markdown(description)

apps/infer.py

@@ -450,16 +450,10 @@ def generate_model(in_path, model_type):
     video_path = os.path.join(config_dict['out_dir'], cfg.name, f"vid/{data['name']}_cloth.mp4")
     overlap_path = os.path.join(config_dict['out_dir'], cfg.name, f"png/{data['name']}_overlap.png")
     
-    del model
-    del dataset
-    del local_affine_model
-    del optimizer_smpl
-    del optimizer_cloth
-    del scheduler_smpl
-    del scheduler_cloth
-    del losses
-    del in_tensor
-    
+    # clean all the variables
+    for element in dir():
+        if 'path' not in element:
+            del locals()[element]
     torch.cuda.empty_cache()
     
     return [smpl_path, smpl_path, smpl_npy_path, recon_path, recon_path, refine_path, refine_path, video_path, overlap_path]