Modelado models, manifest version 1

README.md

@@ -0,0 +1,26 @@
+---
+license: other
+license_name: per-model
+pipeline_tag: image-to-3d
+tags: [onnx, image-to-3d, depth-estimation, triposr, modelado]
+---
+
+# Modelado models (ONNX)
+
+The ONNX models [Modelado](https://github.com/alegauss/modelado) downloads to turn a photo
+into a 3D model on Windows, with DirectML or on the CPU. They are converted from each
+upstream's own weights and checked against PyTorch by `scripts/export_models.py`. Every
+file is listed in `manifest.json` with its size and SHA-256, and the app refuses one that
+differs. Each model keeps its upstream license:
+
+| Model | Task | Variants | License |
+| --- | --- | --- | --- |
+| triposr-encoder | image-to-triplane | fp32 (1675 MB), fp16 (839 MB) | [MIT](https://huggingface.co/stabilityai/TripoSR/blob/5b521936b01fbe1890f6f9baed0254ab6351c04a/README.md) |
+| triposr-decoder | triplane-decoder | fp32 (0.2 MB) | [MIT](https://huggingface.co/stabilityai/TripoSR/blob/5b521936b01fbe1890f6f9baed0254ab6351c04a/README.md) |
+| depth-anything-v2-small | depth | fp32 (99 MB), fp16 (50 MB) | [Apache-2.0](https://huggingface.co/depth-anything/Depth-Anything-V2-Small-hf/blob/5426e4f0f36572d16453bbda7a8389317b1bef99/README.md) |
+| u2net | background-removal | fp32 (176 MB) | [Apache-2.0 (U-2-Net weights), MIT (rembg export)](https://github.com/xuebinqin/U-2-Net/blob/master/LICENSE) |
+| realesrgan-x4plus | upscale | fp32 (67 MB), fp16 (34 MB) | [BSD-3-Clause](https://github.com/xinntao/Real-ESRGAN/blob/master/LICENSE) |
+
+`triposr-encoder` and `triposr-decoder` are [TripoSR](https://huggingface.co/stabilityai/TripoSR)
+split in two: the encoder turns a 512 x 512 photo of an object into a triplane, the decoder
+returns density and colour at any point of it, and the app extracts the mesh with marching cubes.

depth-anything-v2-small-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9afc606a816dfd188f618e0a35cf38368e5b141d4d4519d7e3297a358f2d6a65
+size 49585022

depth-anything-v2-small-fp32.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b3c0e2b2d5f9f70eb747fa2b08734f975221a83a8292052c694dff187f54945
+size 99084949

depth-anything-v2-small-meta.json

@@ -0,0 +1,62 @@
+{
+  "schema": 1,
+  "name": "depth-anything-v2-small",
+  "task": "depth",
+  "inputs": [
+    {
+      "name": "image",
+      "shape": [
+        1,
+        3,
+        518,
+        518
+      ]
+    }
+  ],
+  "outputs": [
+    {
+      "name": "depth",
+      "shape": [
+        1,
+        518,
+        518
+      ]
+    }
+  ],
+  "preprocess": {
+    "size": 518,
+    "value_range": [
+      0.0,
+      1.0
+    ]
+  },
+  "params": {},
+  "validation": {
+    "image": "captured.jpeg",
+    "fp32": {
+      "max_error": 1.4e-05,
+      "psnr": 120.0
+    },
+    "fp16": {
+      "max_error": 0.002632,
+      "psnr": 71.6
+    }
+  },
+  "variants": {
+    "fp32": {
+      "file": "fp32.onnx",
+      "bytes": 99084949,
+      "sha256": "6b3c0e2b2d5f9f70eb747fa2b08734f975221a83a8292052c694dff187f54945"
+    },
+    "fp16": {
+      "file": "fp16.onnx",
+      "bytes": 49585022,
+      "sha256": "9afc606a816dfd188f618e0a35cf38368e5b141d4d4519d7e3297a358f2d6a65"
+    }
+  },
+  "upstream": {
+    "url": "https://huggingface.co/depth-anything/Depth-Anything-V2-Small-hf/resolve/5426e4f0f36572d16453bbda7a8389317b1bef99/model.safetensors",
+    "license": "Apache-2.0",
+    "license_url": "https://huggingface.co/depth-anything/Depth-Anything-V2-Small-hf/blob/5426e4f0f36572d16453bbda7a8389317b1bef99/README.md"
+  }
+}

realesrgan-x4plus-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a2e2f7e3c57ebbe881e0e79f65625298916fe6f0f02265873b2e0ae4e770f81
+size 33651545

realesrgan-x4plus-fp32.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33bad6284772e779db17c7ebb216f25f18be584ba13eb7bc0c0ee16d117ccc0c
+size 67051646

realesrgan-x4plus-meta.json

@@ -0,0 +1,65 @@
+{
+  "schema": 1,
+  "name": "realesrgan-x4plus",
+  "task": "upscale",
+  "inputs": [
+    {
+      "name": "image",
+      "shape": [
+        1,
+        3,
+        "height",
+        "width"
+      ]
+    }
+  ],
+  "outputs": [
+    {
+      "name": "upscaled",
+      "shape": [
+        1,
+        3,
+        "out_height",
+        "out_width"
+      ]
+    }
+  ],
+  "preprocess": {
+    "value_range": [
+      0.0,
+      1.0
+    ]
+  },
+  "params": {
+    "scale": 4,
+    "tile": 256
+  },
+  "validation": {
+    "image": "0014.jpg",
+    "fp32": {
+      "max_error": 1.1e-05,
+      "psnr": 120.0
+    },
+    "fp16": {
+      "max_error": 0.004984,
+      "psnr": 75.87
+    }
+  },
+  "variants": {
+    "fp32": {
+      "file": "fp32.onnx",
+      "bytes": 67051646,
+      "sha256": "33bad6284772e779db17c7ebb216f25f18be584ba13eb7bc0c0ee16d117ccc0c"
+    },
+    "fp16": {
+      "file": "fp16.onnx",
+      "bytes": 33651545,
+      "sha256": "3a2e2f7e3c57ebbe881e0e79f65625298916fe6f0f02265873b2e0ae4e770f81"
+    }
+  },
+  "upstream": {
+    "url": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth",
+    "license": "BSD-3-Clause",
+    "license_url": "https://github.com/xinntao/Real-ESRGAN/blob/master/LICENSE"
+  }
+}

triposr-decoder-fp32.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1604a0dc1903eee485d5a388dd2b4bcac8fad484a96010a797d7c1f40583d985
+size 173374

triposr-decoder-meta.json

@@ -0,0 +1,70 @@
+{
+  "schema": 1,
+  "name": "triposr-decoder",
+  "task": "triplane-decoder",
+  "inputs": [
+    {
+      "name": "triplane",
+      "shape": [
+        1,
+        3,
+        40,
+        64,
+        64
+      ]
+    },
+    {
+      "name": "points",
+      "shape": [
+        "n",
+        3
+      ]
+    }
+  ],
+  "outputs": [
+    {
+      "name": "density",
+      "shape": [
+        "n"
+      ]
+    },
+    {
+      "name": "color",
+      "shape": [
+        "n",
+        3
+      ]
+    }
+  ],
+  "preprocess": {
+    "value_range": [
+      -0.87,
+      0.87
+    ]
+  },
+  "params": {
+    "radius": 0.87,
+    "density_threshold": 25.0,
+    "chunk": 262144
+  },
+  "validation": {
+    "image": "chair.png",
+    "grid": 64,
+    "fp32": {
+      "density_error": 9.3e-05,
+      "color_error": 2.9e-05
+    }
+  },
+  "variants": {
+    "fp32": {
+      "file": "fp32.onnx",
+      "bytes": 173374,
+      "sha256": "1604a0dc1903eee485d5a388dd2b4bcac8fad484a96010a797d7c1f40583d985"
+    }
+  },
+  "upstream": {
+    "url": "https://huggingface.co/stabilityai/TripoSR/resolve/5b521936b01fbe1890f6f9baed0254ab6351c04a/model.ckpt",
+    "license": "MIT",
+    "license_url": "https://huggingface.co/stabilityai/TripoSR/blob/5b521936b01fbe1890f6f9baed0254ab6351c04a/README.md"
+  }
+}

triposr-encoder-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bb475e29ca6f4c936f60cc7e870b7a88350d109c17eb849cc118bcb98051d52
+size 838912052

triposr-encoder-fp32.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:363bcb6cbe44a144ab478a17f29b01af6378e25d81cdc0e04141b69fdb684008
+size 1675325386

triposr-encoder-meta.json

@@ -0,0 +1,70 @@
+{
+  "schema": 1,
+  "name": "triposr-encoder",
+  "task": "image-to-triplane",
+  "inputs": [
+    {
+      "name": "image",
+      "shape": [
+        1,
+        3,
+        512,
+        512
+      ]
+    }
+  ],
+  "outputs": [
+    {
+      "name": "triplane",
+      "shape": [
+        1,
+        3,
+        40,
+        64,
+        64
+      ]
+    }
+  ],
+  "preprocess": {
+    "size": 512,
+    "value_range": [
+      0.0,
+      1.0
+    ],
+    "background": 0.5,
+    "foreground_ratio": 0.85
+  },
+  "params": {
+    "radius": 0.87,
+    "density_threshold": 25.0
+  },
+  "validation": {
+    "image": "chair.png",
+    "grid": 64,
+    "fp32": {
+      "triplane_relative_error": 3e-06,
+      "occupancy_mismatch": 0.0
+    },
+    "fp16": {
+      "triplane_relative_error": 0.000974,
+      "occupancy_mismatch": 0.0
+    }
+  },
+  "variants": {
+    "fp32": {
+      "file": "fp32.onnx",
+      "bytes": 1675325386,
+      "sha256": "363bcb6cbe44a144ab478a17f29b01af6378e25d81cdc0e04141b69fdb684008"
+    },
+    "fp16": {
+      "file": "fp16.onnx",
+      "bytes": 838912052,
+      "sha256": "7bb475e29ca6f4c936f60cc7e870b7a88350d109c17eb849cc118bcb98051d52"
+    }
+  },
+  "upstream": {
+    "url": "https://huggingface.co/stabilityai/TripoSR/resolve/5b521936b01fbe1890f6f9baed0254ab6351c04a/model.ckpt",
+    "license": "MIT",
+    "license_url": "https://huggingface.co/stabilityai/TripoSR/blob/5b521936b01fbe1890f6f9baed0254ab6351c04a/README.md"
+  }
+}

u2net-fp32.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d10d2f3bb75ae3b6d527c77944fc5e7dcd94b29809d47a739a7a728a912b491
+size 175997641

u2net-meta.json

@@ -0,0 +1,64 @@
+{
+  "schema": 1,
+  "name": "u2net",
+  "task": "background-removal",
+  "inputs": [
+    {
+      "name": "input.1",
+      "shape": [
+        1,
+        3,
+        320,
+        320
+      ]
+    }
+  ],
+  "outputs": [
+    {
+      "name": "1959",
+      "shape": [
+        1,
+        1,
+        320,
+        320
+      ]
+    }
+  ],
+  "preprocess": {
+    "size": 320,
+    "value_range": [
+      0.0,
+      1.0
+    ],
+    "scale_by_max": true,
+    "mean": [
+      0.485,
+      0.456,
+      0.406
+    ],
+    "std": [
+      0.229,
+      0.224,
+      0.225
+    ]
+  },
+  "params": {},
+  "validation": {
+    "image": "chair.png",
+    "fp32": {
+      "iou": 0.9786
+    }
+  },
+  "variants": {
+    "fp32": {
+      "file": "fp32.onnx",
+      "bytes": 175997641,
+      "sha256": "8d10d2f3bb75ae3b6d527c77944fc5e7dcd94b29809d47a739a7a728a912b491"
+    }
+  },
+  "upstream": {
+    "url": "https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2net.onnx",
+    "license": "Apache-2.0 (U-2-Net weights), MIT (rembg export)",
+    "license_url": "https://github.com/xuebinqin/U-2-Net/blob/master/LICENSE"
+  }
+}