Image-to-3D
3d
dmitriitochilkin commited on
Commit
3877e74
1 Parent(s): c1cf771

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -5
README.md CHANGED
@@ -14,23 +14,24 @@ pipeline_tag: image-to-3d
14
  ---
15
  # TripoSR
16
  ![](figures/input800.mp4)
17
- TripoSR is a fast and feed-forward 3D generative model developed in partnership with Tripo AI inspired by the recent work of [LRM: Large Reconstruction Model For Single Image to 3D](https://arxiv.org/abs/2311.04400).
18
 
19
  ## Model Details
20
 
21
  ### Model Description
22
 
23
- We closely follow [LRM](https://arxiv.org/abs/2311.04400) network architecture for the model design, where TripoSR incorporates a series of technical advancements over the LRM model in terms of both data curation as well as model and training improvements. For more technical details and evaluations, please refer to [the tech report on our model](link).
24
 
25
  * **Developed by**: [Stability AI](https://stability.ai/), [Tripo AI](https://tripo3d.ai/)
26
- * **Model type**: Feed-forward model
27
  * **License**: MIT
28
- * **Hardware**: `TripoSR` was trained on the Stability AI cluster on 22 nodes with 8 A100 40GBs GPUs for 5 days.
29
 
30
  ### Model Sources
31
 
32
  * **Repository**: https://github.com/VAST-AI-Research/TripoSR
33
  * **Tech report**: *link*
 
34
 
35
  ### Training Dataset
36
 
@@ -39,7 +40,9 @@ We use renders from the [Objaverse](https://objaverse.allenai.org/objaverse-1.0)
39
 
40
  ## Usage
41
 
42
- For usage instructions, please refer to our [TripoSR GitHub repository](https://github.com/VAST-AI-Research/TripoSR).
 
 
43
 
44
 
45
  ### Misuse, Malicious Use, and Out-of-Scope Use
 
14
  ---
15
  # TripoSR
16
  ![](figures/input800.mp4)
17
+ TripoSR is a fast and feed-forward 3D generative model developed in collaboration between Stability AI and Tripo AI.
18
 
19
  ## Model Details
20
 
21
  ### Model Description
22
 
23
+ We closely follow [LRM](https://arxiv.org/abs/2311.04400) network architecture for the model design, where TripoSR incorporates a series of technical advancements over the LRM model in terms of both data curation as well as model and training improvements. For more technical details and evaluations, please refer to [our tech report](link).
24
 
25
  * **Developed by**: [Stability AI](https://stability.ai/), [Tripo AI](https://tripo3d.ai/)
26
+ * **Model type**: Feed-forward 3D reconstruction from a single image
27
  * **License**: MIT
28
+ * **Hardware**: We train `TripoSR` for 5 days on 22 GPU nodes each with 8 A100 40GB GPUs
29
 
30
  ### Model Sources
31
 
32
  * **Repository**: https://github.com/VAST-AI-Research/TripoSR
33
  * **Tech report**: *link*
34
+ * **Demo**: https://huggingface.co/spaces/stabilityai/TripoSR
35
 
36
  ### Training Dataset
37
 
 
40
 
41
  ## Usage
42
 
43
+ * For usage instructions, please refer to our [TripoSR GitHub repository](https://github.com/VAST-AI-Research/TripoSR)
44
+
45
+ * You can also try it in [our gradio demo](https://huggingface.co/spaces/stabilityai/TripoSR)
46
 
47
 
48
  ### Misuse, Malicious Use, and Out-of-Scope Use