camenduru/grm 🔢❓✓🖼️ → 🖼️

▶️ 606 runs 📅 Mar 2024 ⚙️ Cog 0.9.4 🔗 GitHub 📄 Paper ⚖️ License
3d-reconstruction image-to-3d mesh-generation point-cloud video-generation

About

GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation

Example Output

Output

Example outputExample outputExample output

Performance Metrics

25.86s Prediction Time
25.88s Total Time
All Input Parameters
{
  "seed": 21,
  "model": "Zero123++ v1.2",
  "fuse_mesh": true,
  "input_image": "https://replicate.delivery/pbxt/Kf2I8ezAPJ9a6YZJUnDkoGq7urlPtjrA5hRS02D0knxS2KrW/dragon2.png"
}
Input Parameters
seed Type: integerDefault: 42
model Default: Zero123++ v1.2
fuse_mesh Type: booleanDefault: true
input_image (required) Type: string
Input Image
Output Schema

Output

Type: arrayItems Type: stringItems Format: uri

Example Execution Logs
Running image-to-3d with seed 21...
  0%|          | 0/30 [00:00<?, ?it/s]
  7%|▋         | 2/30 [00:00<00:02,  9.99it/s]
 13%|█▎        | 4/30 [00:00<00:02, 10.24it/s]
 20%|██        | 6/30 [00:00<00:02,  9.76it/s]
 27%|██▋       | 8/30 [00:00<00:02, 10.08it/s]
 33%|███▎      | 10/30 [00:00<00:01, 10.21it/s]
 40%|████      | 12/30 [00:01<00:01, 10.23it/s]
 47%|████▋     | 14/30 [00:01<00:01, 10.30it/s]
 53%|█████▎    | 16/30 [00:01<00:01, 10.34it/s]
 60%|██████    | 18/30 [00:01<00:01, 10.03it/s]
 67%|██████▋   | 20/30 [00:01<00:00, 10.11it/s]
 73%|███████▎  | 22/30 [00:02<00:00, 10.19it/s]
 80%|████████  | 24/30 [00:02<00:00, 10.20it/s]
 87%|████████▋ | 26/30 [00:02<00:00, 10.26it/s]
 93%|█████████▎| 28/30 [00:02<00:00,  9.94it/s]
100%|██████████| 30/30 [00:02<00:00, 10.01it/s]
100%|██████████| 30/30 [00:02<00:00, 10.11it/s]
Integrate images into the TSDF volume.
  0%|          | 0/200 [00:00<?, ?it/s]
  0%|          | 1/200 [00:00<00:26,  7.58it/s]
  2%|▏         | 4/200 [00:00<00:10, 18.13it/s]
  4%|▍         | 9/200 [00:00<00:06, 29.80it/s]
  6%|▋         | 13/200 [00:00<00:07, 24.91it/s]
  8%|▊         | 16/200 [00:00<00:07, 25.94it/s]
 10%|▉         | 19/200 [00:00<00:08, 21.92it/s]
 11%|█         | 22/200 [00:01<00:08, 19.84it/s]
 12%|█▎        | 25/200 [00:01<00:08, 20.77it/s]
 15%|█▌        | 30/200 [00:01<00:07, 23.61it/s]
 16%|█▋        | 33/200 [00:01<00:07, 23.33it/s]
 18%|█▊        | 36/200 [00:01<00:06, 24.62it/s]
 20%|█▉        | 39/200 [00:01<00:06, 25.79it/s]
 21%|██        | 42/200 [00:01<00:07, 22.56it/s]
 22%|██▎       | 45/200 [00:01<00:06, 22.93it/s]
 24%|██▍       | 49/200 [00:02<00:06, 23.51it/s]
 26%|██▌       | 52/200 [00:02<00:06, 24.18it/s]
 28%|██▊       | 55/200 [00:02<00:05, 24.87it/s]
 29%|██▉       | 58/200 [00:02<00:05, 25.32it/s]
 31%|███       | 62/200 [00:02<00:05, 25.29it/s]
 32%|███▎      | 65/200 [00:02<00:05, 25.24it/s]
 34%|███▍      | 68/200 [00:02<00:05, 25.60it/s]
 36%|███▌      | 72/200 [00:02<00:04, 28.64it/s]
 38%|███▊      | 76/200 [00:03<00:04, 27.01it/s]
 40%|███▉      | 79/200 [00:03<00:04, 26.65it/s]
 41%|████      | 82/200 [00:03<00:05, 23.33it/s]
 42%|████▎     | 85/200 [00:03<00:04, 24.28it/s]
 44%|████▍     | 88/200 [00:03<00:04, 24.93it/s]
 46%|████▌     | 92/200 [00:03<00:03, 28.13it/s]
 48%|████▊     | 96/200 [00:03<00:03, 30.36it/s]
 50%|█████     | 100/200 [00:04<00:03, 27.14it/s]
 52%|█████▏    | 104/200 [00:04<00:03, 29.18it/s]
 54%|█████▍    | 108/200 [00:04<00:03, 26.40it/s]
 56%|█████▌    | 111/200 [00:04<00:03, 26.15it/s]
 57%|█████▋    | 114/200 [00:04<00:03, 26.57it/s]
 59%|█████▉    | 118/200 [00:04<00:03, 26.03it/s]
 60%|██████    | 121/200 [00:04<00:03, 26.25it/s]
 62%|██████▎   | 125/200 [00:04<00:02, 28.84it/s]
 64%|██████▍   | 129/200 [00:05<00:02, 31.05it/s]
 66%|██████▋   | 133/200 [00:05<00:02, 27.33it/s]
 68%|██████▊   | 136/200 [00:05<00:02, 27.21it/s]
 70%|███████   | 140/200 [00:05<00:02, 29.15it/s]
 72%|███████▏  | 143/200 [00:05<00:02, 25.60it/s]
 73%|███████▎  | 146/200 [00:05<00:02, 25.64it/s]
 75%|███████▌  | 150/200 [00:05<00:01, 28.53it/s]
 76%|███████▋  | 153/200 [00:05<00:01, 28.36it/s]
 78%|███████▊  | 156/200 [00:06<00:01, 24.70it/s]
 80%|████████  | 160/200 [00:06<00:01, 27.08it/s]
 82%|████████▏ | 163/200 [00:06<00:01, 26.99it/s]
 83%|████████▎ | 166/200 [00:06<00:01, 27.05it/s]
 84%|████████▍ | 169/200 [00:06<00:01, 24.02it/s]
 87%|████████▋ | 174/200 [00:06<00:00, 29.33it/s]
 89%|████████▉ | 178/200 [00:06<00:00, 26.24it/s]
 91%|█████████ | 182/200 [00:07<00:00, 28.56it/s]
 94%|█████████▎| 187/200 [00:07<00:00, 32.72it/s]
 96%|█████████▌| 191/200 [00:07<00:00, 29.52it/s]
 98%|█████████▊| 196/200 [00:07<00:00, 33.36it/s]
100%|██████████| 200/200 [00:07<00:00, 26.59it/s]
Extract a triangle mesh from the volume and export it.
[Open3D DEBUG] [ClusterConnectedTriangles] Compute triangle adjacency
[Open3D DEBUG] [ClusterConnectedTriangles] Done computing triangle adjacency
[Open3D DEBUG] [ClusterConnectedTriangles] Done clustering, #clusters=46060
[Mesh loading] v: torch.Size([70902, 3]), f: torch.Size([141445, 3])
[Mesh loading] vn: torch.Size([70902, 3]), fn: torch.Size([141445, 3])
tensor(0.2557, device='cuda:0') tensor(0.7722, device='cuda:0')
Version Details
Version ID
8c489b23ddc6a9b0484f8b6f55f9b24fd792a84a6715d02f7e3ec63325bae4f0
Version Created
March 30, 2024
Run on Replicate →