| Name | lvsm-pytorch JSON |
| Version |
0.0.12
JSON |
| download |
| home_page | None |
| Summary | LVSM - Pytorch |
| upload_time | 2024-11-05 18:55:55 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.9 |
| license | MIT License Copyright (c) 2024 Phil Wang Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. |
| keywords |
artificial intelligence
attention mechanism
deep learning
novel view synthesis
transformers
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
<img src="./lvsm.png" width="500px"></img>
<img src="./lvsm-finding.png" width="400px"></img>
<img src="./plucker-ray.png" width="400px"></img>
## LVSM - Pytorch (wip)
Implementation of [LVSM](https://haian-jin.github.io/projects/LVSM/), SOTA Large View Synthesis with Minimal 3d Inductive Bias, from Adobe Research
We will focus only on the Decoder-only architecture in this repository.
This paper lines up with <a href="https://openreview.net/forum?id=A8Vuf2e8y6">another</a> from ICLR 2025
## Install
```bash
$ pip install lvsm-pytorch
```
## Usage
```python
import torch
from lvsm_pytorch import LVSM
rays = torch.randn(2, 4, 6, 256, 256)
images = torch.randn(2, 4, 3, 256, 256)
target_rays = torch.randn(2, 6, 256, 256)
target_images = torch.randn(2, 3, 256, 256)
model = LVSM(
dim = 512,
max_image_size = 256,
patch_size = 32,
depth = 2,
)
loss = model(
input_images = images,
input_rays = rays,
target_rays = target_rays,
target_images = target_images
)
loss.backward()
# after much training
pred_images = model(
input_images = images,
input_rays = rays,
target_rays = target_rays,
) # (2, 3, 256, 256)
assert pred_images.shape == target_images.shape
```
Or from the raw camera intrinsic / extrinsics (please submit an issue or pull request if you see an error. new to view synthesis and out of my depths here)
```python
import torch
from lvsm_pytorch.lvsm import LVSM, CameraWrapper
input_intrinsic_rotation = torch.randn(2, 4, 3, 3)
input_extrinsic_rotation = torch.randn(2, 4, 3, 3)
input_translation = torch.randn(2, 4, 3)
input_uniform_points = torch.randn(2, 4, 3, 256, 256)
target_intrinsic_rotation = torch.randn(2, 3, 3)
target_extrinsic_rotation = torch.randn(2, 3, 3)
target_translation = torch.randn(2, 3)
target_uniform_points = torch.randn(2, 3, 256, 256)
images = torch.randn(2, 4, 4, 256, 256)
target_images = torch.randn(2, 4, 256, 256)
lvsm = LVSM(
dim = 512,
max_image_size = 256,
patch_size = 32,
channels = 4,
depth = 2,
)
model = CameraWrapper(lvsm)
loss = model(
input_intrinsic_rotation = input_intrinsic_rotation,
input_extrinsic_rotation = input_extrinsic_rotation,
input_translation = input_translation,
input_uniform_points = input_uniform_points,
target_intrinsic_rotation = target_intrinsic_rotation,
target_extrinsic_rotation = target_extrinsic_rotation,
target_translation = target_translation,
target_uniform_points = target_uniform_points,
input_images = images,
target_images = target_images,
)
loss.backward()
```
## Citations
```bibtex
@inproceedings{Jin2024LVSMAL,
title = {LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias},
author = {Haian Jin and Hanwen Jiang and Hao Tan and Kai Zhang and Sai Bi and Tianyuan Zhang and Fujun Luan and Noah Snavely and Zexiang Xu},
year = {2024},
url = {https://api.semanticscholar.org/CorpusID:273507016}
}
```
```bibtex
@article{Zhang2024CamerasAR,
title = {Cameras as Rays: Pose Estimation via Ray Diffusion},
author = {Jason Y. Zhang and Amy Lin and Moneish Kumar and Tzu-Hsuan Yang and Deva Ramanan and Shubham Tulsiani},
journal = {ArXiv},
year = {2024},
volume = {abs/2402.14817},
url = {https://api.semanticscholar.org/CorpusID:267782978}
}
```
Raw data
{
"_id": null,
"home_page": null,
"name": "lvsm-pytorch",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "artificial intelligence, attention mechanism, deep learning, novel view synthesis, transformers",
"author": null,
"author_email": "Phil Wang <lucidrains@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/84/15/706e69388e4ac061944842ef73fb50a84d025e5e0fdceedbb20a2678198d/lvsm_pytorch-0.0.12.tar.gz",
"platform": null,
"description": "<img src=\"./lvsm.png\" width=\"500px\"></img>\n\n<img src=\"./lvsm-finding.png\" width=\"400px\"></img>\n\n<img src=\"./plucker-ray.png\" width=\"400px\"></img>\n\n## LVSM - Pytorch (wip)\n\nImplementation of [LVSM](https://haian-jin.github.io/projects/LVSM/), SOTA Large View Synthesis with Minimal 3d Inductive Bias, from Adobe Research\n\nWe will focus only on the Decoder-only architecture in this repository.\n\nThis paper lines up with <a href=\"https://openreview.net/forum?id=A8Vuf2e8y6\">another</a> from ICLR 2025\n\n## Install\n\n```bash\n$ pip install lvsm-pytorch\n```\n\n## Usage\n\n```python\nimport torch\nfrom lvsm_pytorch import LVSM\n\nrays = torch.randn(2, 4, 6, 256, 256)\nimages = torch.randn(2, 4, 3, 256, 256)\n\ntarget_rays = torch.randn(2, 6, 256, 256)\ntarget_images = torch.randn(2, 3, 256, 256)\n\nmodel = LVSM(\n dim = 512,\n max_image_size = 256,\n patch_size = 32,\n depth = 2,\n)\n\nloss = model(\n input_images = images,\n input_rays = rays,\n target_rays = target_rays,\n target_images = target_images\n)\n\nloss.backward()\n\n# after much training\n\npred_images = model(\n input_images = images,\n input_rays = rays,\n target_rays = target_rays,\n) # (2, 3, 256, 256)\n\nassert pred_images.shape == target_images.shape\n```\n\nOr from the raw camera intrinsic / extrinsics (please submit an issue or pull request if you see an error. new to view synthesis and out of my depths here)\n\n```python\nimport torch\nfrom lvsm_pytorch.lvsm import LVSM, CameraWrapper\n\ninput_intrinsic_rotation = torch.randn(2, 4, 3, 3)\ninput_extrinsic_rotation = torch.randn(2, 4, 3, 3)\ninput_translation = torch.randn(2, 4, 3)\ninput_uniform_points = torch.randn(2, 4, 3, 256, 256)\n\ntarget_intrinsic_rotation = torch.randn(2, 3, 3)\ntarget_extrinsic_rotation = torch.randn(2, 3, 3)\ntarget_translation = torch.randn(2, 3)\ntarget_uniform_points = torch.randn(2, 3, 256, 256)\n\nimages = torch.randn(2, 4, 4, 256, 256)\ntarget_images = torch.randn(2, 4, 256, 256)\n\nlvsm = LVSM(\n dim = 512,\n max_image_size = 256,\n patch_size = 32,\n channels = 4,\n depth = 2,\n)\n\nmodel = CameraWrapper(lvsm)\n\nloss = model(\n input_intrinsic_rotation = input_intrinsic_rotation,\n input_extrinsic_rotation = input_extrinsic_rotation,\n input_translation = input_translation,\n input_uniform_points = input_uniform_points,\n target_intrinsic_rotation = target_intrinsic_rotation,\n target_extrinsic_rotation = target_extrinsic_rotation,\n target_translation = target_translation,\n target_uniform_points = target_uniform_points,\n input_images = images,\n target_images = target_images,\n)\n\nloss.backward()\n```\n\n## Citations\n\n```bibtex\n@inproceedings{Jin2024LVSMAL,\n title = {LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias},\n author = {Haian Jin and Hanwen Jiang and Hao Tan and Kai Zhang and Sai Bi and Tianyuan Zhang and Fujun Luan and Noah Snavely and Zexiang Xu},\n year = {2024},\n url = {https://api.semanticscholar.org/CorpusID:273507016}\n}\n```\n\n```bibtex\n@article{Zhang2024CamerasAR,\n title = {Cameras as Rays: Pose Estimation via Ray Diffusion},\n author = {Jason Y. Zhang and Amy Lin and Moneish Kumar and Tzu-Hsuan Yang and Deva Ramanan and Shubham Tulsiani},\n journal = {ArXiv},\n year = {2024},\n volume = {abs/2402.14817},\n url = {https://api.semanticscholar.org/CorpusID:267782978}\n}\n```\n",
"bugtrack_url": null,
"license": "MIT License Copyright (c) 2024 Phil Wang Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the \"Software\"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.",
"summary": "LVSM - Pytorch",
"version": "0.0.12",
"project_urls": {
"Homepage": "https://pypi.org/project/lvsm-pytorch/",
"Repository": "https://github.com/lucidrains/lvsm-pytorch"
},
"split_keywords": [
"artificial intelligence",
" attention mechanism",
" deep learning",
" novel view synthesis",
" transformers"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "d4f682aa4f4d2ec5e965605c8220e3203a207f50f1f8c8c91fdb417f494a7eba",
"md5": "dd94679dd5c2134ec7e25859a62fe380",
"sha256": "6c0eb09783d0d64a45ee4cec021e2645739335b64269b39c477d309878bda9ec"
},
"downloads": -1,
"filename": "lvsm_pytorch-0.0.12-py3-none-any.whl",
"has_sig": false,
"md5_digest": "dd94679dd5c2134ec7e25859a62fe380",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 7254,
"upload_time": "2024-11-05T18:55:54",
"upload_time_iso_8601": "2024-11-05T18:55:54.348745Z",
"url": "https://files.pythonhosted.org/packages/d4/f6/82aa4f4d2ec5e965605c8220e3203a207f50f1f8c8c91fdb417f494a7eba/lvsm_pytorch-0.0.12-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "8415706e69388e4ac061944842ef73fb50a84d025e5e0fdceedbb20a2678198d",
"md5": "f6310ef1c24980ff4eba1bf0b6a19a46",
"sha256": "ff497bc62b27669b2031e9fbb8aa13a9db3b03fa458d2eb6cebed175386a4584"
},
"downloads": -1,
"filename": "lvsm_pytorch-0.0.12.tar.gz",
"has_sig": false,
"md5_digest": "f6310ef1c24980ff4eba1bf0b6a19a46",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 1489418,
"upload_time": "2024-11-05T18:55:55",
"upload_time_iso_8601": "2024-11-05T18:55:55.665094Z",
"url": "https://files.pythonhosted.org/packages/84/15/706e69388e4ac061944842ef73fb50a84d025e5e0fdceedbb20a2678198d/lvsm_pytorch-0.0.12.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-11-05 18:55:55",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "lucidrains",
"github_project": "lvsm-pytorch",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "lvsm-pytorch"
}
