d4rt 0.0.1__tar.gz

d4rt-0.0.1/.gitignore

@@ -0,0 +1,220 @@
+*.sh
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[codz]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#   Usually these files are written by a python script from a template
+#   before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py.cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+# Pipfile.lock
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+# uv.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+# poetry.lock
+# poetry.toml
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#   pdm recommends including project-wide configuration in pdm.toml, but excluding .pdm-python.
+#   https://pdm-project.org/en/latest/usage/project/#working-with-version-control
+# pdm.lock
+# pdm.toml
+.pdm-python
+.pdm-build/
+
+# pixi
+#   Similar to Pipfile.lock, it is generally recommended to include pixi.lock in version control.
+# pixi.lock
+#   Pixi creates a virtual environment in the .pixi directory, just like venv module creates one
+#   in the .venv directory. It is recommended not to include this directory in version control.
+.pixi
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# Redis
+*.rdb
+*.aof
+*.pid
+
+# RabbitMQ
+mnesia/
+rabbitmq/
+rabbitmq-data/
+
+# ActiveMQ
+activemq-data/
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.envrc
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#   JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#   be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#   and can be added to the global gitignore or merged into this file.  For a more nuclear
+#   option (not recommended) you can uncomment the following to ignore the entire idea folder.
+# .idea/
+
+# Abstra
+#   Abstra is an AI-powered process automation framework.
+#   Ignore directories containing user credentials, local state, and settings.
+#   Learn more at https://abstra.io/docs
+.abstra/
+
+# Visual Studio Code
+#   Visual Studio Code specific template is maintained in a separate VisualStudioCode.gitignore 
+#   that can be found at https://github.com/github/gitignore/blob/main/Global/VisualStudioCode.gitignore
+#   and can be added to the global gitignore or merged into this file. However, if you prefer, 
+#   you could uncomment the following to ignore the entire vscode folder
+# .vscode/
+# Temporary file for partial code execution
+tempCodeRunnerFile.py
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# Marimo
+marimo/_static/
+marimo/_lsp/
+__marimo__/
+
+# Streamlit
+.streamlit/secrets.toml

d4rt-0.0.1/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 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.

d4rt-0.0.1/PKG-INFO

@@ -0,0 +1,72 @@
+Metadata-Version: 2.4
+Name: d4rt
+Version: 0.0.1
+Summary: Implementation of D4RT, Efficiently Reconstructing Dynamic Scenes
+Project-URL: Homepage, https://pypi.org/project/d4rt/
+Project-URL: Repository, https://codeberg.org/lucidrains/d4rt
+Author-email: Phil Wang <lucidrains@gmail.com>
+License: MIT License
+        
+        Copyright (c) 2026 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.
+License-File: LICENSE
+Keywords: 4d,artificial intelligence,deep learning
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Requires-Dist: einops>=0.8.1
+Requires-Dist: einx>=0.4.3
+Requires-Dist: torch-einops-utils>=0.0.32
+Requires-Dist: torch>=2.5
+Requires-Dist: x-transformers>=2.18.10
+Provides-Extra: examples
+Provides-Extra: test
+Requires-Dist: pytest; extra == 'test'
+Description-Content-Type: text/markdown
+
+<img src="./d4rt.png" width="400px"></img>
+
+## d4rt (wip)
+
+Implementation of [D4RT](https://d4rt-paper.github.io/), Efficiently Reconstructing Dynamic Scenes, Deepmind
+
+## citations
+
+```bibtex
+@article{zhang2025d4rt,
+    title   = {Efficiently Reconstructing Dynamic Scenes One D4RT at a Time},
+    author  = {Zhang, Chuhan and Le Moing, Guillaume and Koppula, Skanda and Rocco, Ignacio and Momeni, Liliane and Xie, Junyu and Sun, Shuyang and Sukthankar, Rahul and Barral, Jo{\"e}lle K. and Hadsell, Raia and Ghahramani, Zoubin and Zisserman, Andrew and Zhang, Junlin and Sajjadi, Mehdi S. M.},
+    journal = {arXiv preprint},
+    year    = {2025}
+}
+```
+
+```bibtex
+@inproceedings{liu2026geometryaware,
+    title   = {Geometry-aware 4D Video Generation for Robot Manipulation},
+    author  = {Zeyi Liu and Shuang Li and Eric Cousineau and Siyuan Feng and Benjamin Burchfiel and Shuran Song},
+    booktitle = {The Fourteenth International Conference on Learning Representations},
+    year    = {2026},
+    url     = {https://openreview.net/forum?id=18gC6pZVVc}
+}
+```

d4rt-0.0.1/README.md

@@ -0,0 +1,26 @@
+<img src="./d4rt.png" width="400px"></img>
+
+## d4rt (wip)
+
+Implementation of [D4RT](https://d4rt-paper.github.io/), Efficiently Reconstructing Dynamic Scenes, Deepmind
+
+## citations
+
+```bibtex
+@article{zhang2025d4rt,
+    title   = {Efficiently Reconstructing Dynamic Scenes One D4RT at a Time},
+    author  = {Zhang, Chuhan and Le Moing, Guillaume and Koppula, Skanda and Rocco, Ignacio and Momeni, Liliane and Xie, Junyu and Sun, Shuyang and Sukthankar, Rahul and Barral, Jo{\"e}lle K. and Hadsell, Raia and Ghahramani, Zoubin and Zisserman, Andrew and Zhang, Junlin and Sajjadi, Mehdi S. M.},
+    journal = {arXiv preprint},
+    year    = {2025}
+}
+```
+
+```bibtex
+@inproceedings{liu2026geometryaware,
+    title   = {Geometry-aware 4D Video Generation for Robot Manipulation},
+    author  = {Zeyi Liu and Shuang Li and Eric Cousineau and Siyuan Feng and Benjamin Burchfiel and Shuran Song},
+    booktitle = {The Fourteenth International Conference on Learning Representations},
+    year    = {2026},
+    url     = {https://openreview.net/forum?id=18gC6pZVVc}
+}
+```

d4rt-0.0.1/d4rt/__init__.py

@@ -0,0 +1 @@
+from d4rt.d4rt import D4RT

d4rt-0.0.1/d4rt/d4rt.py

@@ -0,0 +1,167 @@
+from __future__ import annotations
+
+import torch
+from torch import nn
+import torch.nn.functional as F
+from torch.nn import Module, ModuleList, Sequential
+
+from x_transformers import Encoder, CrossAttender, Attention, FeedForward
+
+# ein notation
+
+import einx
+from einops import rearrange
+from einops.layers.torch import Rearrange
+from torch_einops_utils import pack_with_inverse
+
+# helpers
+
+def exists(v):
+    return v is not None
+
+# video self attention encoder
+
+class VideoEncoder(Module):
+    def __init__(
+        self,
+        *,
+        dim,
+        depth,
+        image_size,
+        patch_size,
+        max_time_len,
+        channels = 3,
+        dim_head = 64,
+        heads = 8,
+        ff_glu = True,
+        attn_kwargs: dict = dict(),
+        ff_kwargs: dict = dict()
+    ):
+        super().__init__()
+
+        dim_patch = channels * patch_size * patch_size
+
+        self.patch_to_tokens = Sequential(
+            Rearrange('b t c (h p1) (w p2) -> b t (h w) (p1 p2 c)', p1 = patch_size, p2 = patch_size),
+            nn.Linear(dim_patch, dim),
+            nn.LayerNorm(dim, bias = False)
+        )
+
+        self.layers = ModuleList([])
+
+        for _ in range(depth):
+
+            spatial_attn = Attention(dim = dim, dim_head = dim_head, heads = heads, **attn_kwargs)
+
+            time_attn = Attention(dim = dim, dim_head = dim_head, heads = heads, **attn_kwargs)
+
+            ff = FeedForward(dim = dim, glu = ff_glu, **ff_kwargs)
+
+            self.layers.append(ModuleList([spatial_attn, time_attn, ff]))
+
+        self.norm = nn.LayerNorm(dim, bias = False)
+
+    def forward(
+        self,
+        video # float[b t c h w]
+    ): # float[b n d]
+
+        tokens = self.patch_to_tokens(video) # float[b t s d]
+
+        for spatial_attn, time_attn, ff in self.layers:
+
+            # space attn
+
+            tokens, inverse_pack = pack_with_inverse(tokens, '* s d')
+
+            tokens = spatial_attn(tokens) + tokens
+
+            tokens = inverse_pack(tokens)
+
+            # time attn
+
+            tokens = rearrange(tokens, 'b t s d -> b s t d')
+
+            tokens, inverse_pack = pack_with_inverse(tokens, '* t d')
+
+            tokens = time_attn(tokens) + tokens
+
+            tokens = inverse_pack(tokens)
+
+            tokens = rearrange(tokens, 'b s t d -> b t s d')
+
+            # feedforward
+
+            tokens = ff(tokens) + tokens
+
+        return self.norm(tokens)
+
+# main class
+
+class D4RT(Module):
+    def __init__(
+        self,
+        *,
+        dim,
+        video_image_size,
+        video_patch_size,
+        video_max_time_len,
+        enc_depth,
+        dec_depth,
+        enc_dim_head = 64,
+        enc_heads = 8,
+        dec_dim_head = 64,
+        dec_heads = 8,
+        video_enc_attn_kwargs: dict = dict(),
+        video_enc_ff_kwargs: dict = dict(),
+        cross_attender_kwargs: dict = dict()
+    ):
+        super().__init__()
+
+        self.to_global_spatial_repr = VideoEncoder(
+            dim = dim,
+            depth = enc_depth,
+            dim_head = enc_dim_head,
+            heads = enc_heads,
+            image_size = video_image_size,
+            patch_size = video_patch_size,
+            max_time_len = video_max_time_len,
+            attn_kwargs = video_enc_attn_kwargs,
+            ff_kwargs = video_enc_ff_kwargs
+        )
+
+        self.cross_attender = CrossAttender(
+            dim = dim,
+            depth = dec_depth,
+            heads = dec_heads,
+            attn_dim_head = dec_dim_head,
+            **cross_attender_kwargs
+        )
+
+        self.to_pred = nn.Linear(dim, 3, bias = False)
+
+    def forward(
+        self,
+        video,              # float[b t c h w]
+        queries,            # float[b q d]
+        points = None,      # float[b q 3]
+        return_pred = False
+    ):
+
+        global_spatial_repr = self.to_global_spatial_repr(video)
+
+        global_spatial_repr, inverse_pack_spacetime = pack_with_inverse(global_spatial_repr, 'b * d')
+
+        queried = self.cross_attender(queries, context = global_spatial_repr)
+
+        pred = self.to_pred(queried)
+
+        if not exists(points):
+            return pred
+
+        loss = F.mse_loss(pred, points)
+
+        if not return_pred:
+            return loss
+
+        return loss, pred

d4rt-0.0.1/pyproject.toml

@@ -0,0 +1,68 @@
+[project]
+name = "d4rt"
+version = "0.0.1"
+description = "Implementation of D4RT, Efficiently Reconstructing Dynamic Scenes"
+authors = [
+    { name = "Phil Wang", email = "lucidrains@gmail.com" }
+]
+readme = "README.md"
+requires-python = ">= 3.10"
+license = { file = "LICENSE" }
+keywords = [
+    'artificial intelligence',
+    'deep learning',
+    '4d'
+]
+
+classifiers=[
+    'Development Status :: 4 - Beta',
+    'Intended Audience :: Developers',
+    'Topic :: Scientific/Engineering :: Artificial Intelligence',
+    'License :: OSI Approved :: MIT License',
+    'Programming Language :: Python :: 3.10',
+]
+
+dependencies = [
+    "einx>=0.4.3",
+    "einops>=0.8.1",
+    "torch>=2.5",
+    "torch-einops-utils>=0.0.32",
+    "x-transformers>=2.18.10"
+]
+
+[project.urls]
+Homepage = "https://pypi.org/project/d4rt/"
+Repository = "https://codeberg.org/lucidrains/d4rt"
+
+[project.optional-dependencies]
+examples = []
+test = [
+    "pytest"
+]
+
+[tool.pytest.ini_options]
+pythonpath = [
+    "."
+]
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.rye]
+managed = true
+dev-dependencies = []
+
+[tool.hatch.metadata]
+allow-direct-references = true
+
+[tool.hatch.build]
+include = [
+    "d4rt",
+    "pyproject.toml",
+    "README.md",
+    "LICENSE"
+]
+
+[tool.hatch.build.targets.wheel]
+packages = ["d4rt"]