amiedd commited on
Commit
3641689
·
1 Parent(s): 95a5812

button changes

Browse files
Files changed (4) hide show
  1. LICENSE +1 -1
  2. README.md +2 -3
  3. app.py +9 -3
  4. datasets/README.md +2 -3
LICENSE CHANGED
@@ -1,6 +1,6 @@
1
  MIT License
2
 
3
- Copyright (c) 2025 Yuchen Lin
4
 
5
  Permission is hereby granted, free of charge, to any person obtaining a copy
6
  of this software and associated documentation files (the "Software"), to deal
 
1
  MIT License
2
 
3
+ Copyright (c) 2025 AmieDD
4
 
5
  Permission is hereby granted, free of charge, to any person obtaining a copy
6
  of this software and associated documentation files (the "Software"), to deal
README.md CHANGED
@@ -1,5 +1,5 @@
1
  ---
2
- title: PartCrafter
3
  emoji: 🧩
4
  colorFrom: blue
5
  colorTo: pink
@@ -8,7 +8,6 @@ sdk_version: 5.36.2
8
  app_file: app.py
9
  pinned: false
10
  license: mit
11
- short_description: 3D Mesh Generation via Compositional Latent Diffusion
12
  ---
13
 
14
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
1
  ---
2
+ title: 3D PartAlchemy
3
  emoji: 🧩
4
  colorFrom: blue
5
  colorTo: pink
 
8
  app_file: app.py
9
  pinned: false
10
  license: mit
11
+ short_description: Image to 3D Mesh Generation
12
  ---
13
 
 
app.py CHANGED
@@ -298,6 +298,14 @@ def build_demo():
298
  margin: 0 auto;
299
  max-width: 1280px;
300
  }
 
 
 
 
 
 
 
 
301
  """
302
  theme = gr.themes.Ocean()
303
 
@@ -321,12 +329,10 @@ def build_demo():
321
  with gr.Row():
322
  with gr.Column(scale=1):
323
  gr.Markdown(
324
- """
325
- • We would like to acknowledge : [@alexandernasa](https://twitter.com/alexandernasa/) for the contribution of the Hugging Face Space. """
326
  )
327
  input_image = gr.Image(type="filepath", label="Input Image", height=256)
328
  num_parts = gr.Slider(1, MAX_NUM_PARTS, value=4, step=1, label="Number of Parts")
329
- run_button = gr.Button("🧩 Generate 3D Parts", variant="primary")
330
 
331
  with gr.Accordion("Advanced Settings", open=False):
332
  seed = gr.Number(value=0, label="Random Seed", precision=0)
 
298
  margin: 0 auto;
299
  max-width: 1280px;
300
  }
301
+ .pink-button {
302
+ background: linear-gradient(45deg, #ff69b4, #ff1493) !important;
303
+ border: none !important;
304
+ color: white !important;
305
+ }
306
+ .pink-button:hover {
307
+ background: linear-gradient(45deg, #ff1493, #dc143c) !important;
308
+ }
309
  """
310
  theme = gr.themes.Ocean()
311
 
 
329
  with gr.Row():
330
  with gr.Column(scale=1):
331
  gr.Markdown(
 
 
332
  )
333
  input_image = gr.Image(type="filepath", label="Input Image", height=256)
334
  num_parts = gr.Slider(1, MAX_NUM_PARTS, value=4, step=1, label="Number of Parts")
335
+ run_button = gr.Button("🧩 Generate 3D Parts", variant="primary", elem_classes=["pink-button"])
336
 
337
  with gr.Accordion("Advanced Settings", open=False):
338
  seed = gr.Number(value=0, label="Random Seed", precision=0)
datasets/README.md CHANGED
@@ -1,8 +1,7 @@
1
- # Dataset Preparation
2
- We provide the data preprocessing pipeline for PartCrafter. By following the instructions, you can generate the training data from the raw GLB data. While we are considering releasing the preprocessed dataset, please note that it may take some time before it becomes available.
3
 
4
  ## Download Raw Data
5
- Our final model uses a subset of [Objaverse](https://huggingface.co/datasets/allenai/objaverse) provided by [LGM](https://github.com/ashawkey/objaverse_filter) and [Amazon Berkeley Objects (ABO) Dataset](https://amazon-berkeley-objects.s3.amazonaws.com/index.html). Please download the raw GLB files according to their instructions. You can also use other source of data.
6
 
7
  ## Data Preprocess
8
  We provide several scripts to preprocess the raw GLB files [here](./preprocess/). These scripts are minimal implementations and illustrate the whole preprocessing pipeline on a single 3D object.
 
1
+
 
2
 
3
  ## Download Raw Data
4
+ Our final model uses a subset of [Objaverse](https://huggingface.co/datasets/allenai/objaverse) provided by [LGM](https://github.com/ashawkey/objaverse_filter)
5
 
6
  ## Data Preprocess
7
  We provide several scripts to preprocess the raw GLB files [here](./preprocess/). These scripts are minimal implementations and illustrate the whole preprocessing pipeline on a single 3D object.