Linzhan commited on
Commit
b127ca4
·
verified ·
1 Parent(s): c0cb733

Point at the open data_process pipeline that turns this raw layer into UniML3D

Browse files
Files changed (1) hide show
  1. README.md +14 -9
README.md CHANGED
@@ -1,12 +1,12 @@
1
  ---
 
 
 
 
2
  language:
3
  - en
4
- license: other
5
  size_categories:
6
  - 1K<n<10K
7
- pretty_name: Truebones ZOO Annotations
8
- license_name: truebones-terms-of-use
9
- license_link: https://truebones.gumroad.com/l/skZMC
10
  tags:
11
  - 3d
12
  - animation
@@ -24,14 +24,10 @@ configs:
24
  data_files: species.csv
25
  - config_name: clips
26
  data_files: clips.csv
27
- task_categories:
28
- - other
29
  ---
30
 
31
  # Truebones ZOO Annotations
32
 
33
- [Project page](https://linzhanmou.com/unimate/) | [Code](https://github.com/Friedrich-M/UniMate)
34
-
35
  Text prompts, per-clip metadata, rest-pose renders and the exact build pipeline for
36
  **Truebones ZOO** — **1,097 animal motion clips across 74 skeletons**: mammals, birds,
37
  reptiles, insects, marine and prehistoric creatures. 1.02 hours, 111,158 frames, uniformly
@@ -226,6 +222,15 @@ CC-BY-NC-4.0 — non-commercial use only, narrower than the rest of this reposit
226
  captions are shared by two clips each; retrieval metrics should group ground truth by caption
227
  text so those pairs are not scored as misses.
228
 
 
 
 
 
 
 
 
 
 
229
  ## Usage
230
 
231
  ```python
@@ -292,4 +297,4 @@ This dataset is part of **UniML3D**, the training corpus introduced in
292
  journal = {arXiv preprint arXiv:2609.05415},
293
  year = {2026}
294
  }
295
- ```
 
1
  ---
2
+ license: other
3
+ license_name: truebones-terms-of-use
4
+ license_link: https://truebones.gumroad.com/l/skZMC
5
+ pretty_name: Truebones ZOO Annotations
6
  language:
7
  - en
 
8
  size_categories:
9
  - 1K<n<10K
 
 
 
10
  tags:
11
  - 3d
12
  - animation
 
24
  data_files: species.csv
25
  - config_name: clips
26
  data_files: clips.csv
 
 
27
  ---
28
 
29
  # Truebones ZOO Annotations
30
 
 
 
31
  Text prompts, per-clip metadata, rest-pose renders and the exact build pipeline for
32
  **Truebones ZOO** — **1,097 animal motion clips across 74 skeletons**: mammals, birds,
33
  reptiles, insects, marine and prehistoric creatures. 1.02 hours, 111,158 frames, uniformly
 
222
  captions are shared by two clips each; retrieval metrics should group ground truth by caption
223
  text so those pairs are not scored as misses.
224
 
225
+ ## Processing it into motion data
226
+
227
+ This repository is the **raw** layer. The pipeline that turns it into canonicalized, text-paired motion clips is open at [`UniMate/data_process`](https://github.com/Friedrich-M/UniMate/tree/main/data_process): it exports each clip to NPZ, renders previews, captions them with a vision-language model, cleans the joint labels and derives the training features. The processed release built from it is [UniML3D](https://huggingface.co/datasets/Linzhan/UniML3D).
228
+
229
+ ```bash
230
+ bash data_process/scripts/run_download.sh truebones
231
+ bash data_process/scripts/run_export.sh truebones
232
+ ```
233
+
234
  ## Usage
235
 
236
  ```python
 
297
  journal = {arXiv preprint arXiv:2609.05415},
298
  year = {2026}
299
  }
300
+ ```