Direct3D-S2

Direct3D-S2 is DreamTech's image-to-3D model. It generates a signed distance field on sparse voxels, first at 512^3 and then at 1024^3, and returns the object as a mesh. Upstream needs CUDA-only extensions; LibreYOLO runs the same weights in plain PyTorch through the LibreGen factory.

Zadania
3d generation
Rozmiary
v11 at 518 px
Instalacja
pip install "libreyolo[gen]"
Poziom obsługi
Poziom siostrzany, od wersji v1.8.0. Osobna część produktu z własną fabryką i kontraktem.
Projekt źródłowy
Direct3D-S2, autorzy: DreamTech, licencja: MIT. Publikacja, kod źródłowy
Licencje
Kod: MIT and Apache-2.0, wagi: MIT. Użycie komercyjne

Install

bash
pip install "libreyolo[gen]"

The extra brings diffusers, transformers (DINOv2 encoder) and scikit-image (marching cubes).

Predict

Nothing is gated and no login is needed. Weights download on first use at pinned revisions.

Python
from libreyolo import LibreGen, SAMPLE_IMAGE model = LibreGen("direct3d-s2")     # weights download on first useresult = model(SAMPLE_IMAGE, seed=0) result.save("object.obj")m = result.meshesprint(m.vertices[0].shape, m.faces[0].shape)
CLI
# Saves the object as .obj under runs/generate/predict.libreyolo generate source=https://raw.githubusercontent.com/LibreYOLO/libreyolo/release/libreyolo/assets/guggenheim-bilbao.jpg model=direct3d-s2 seed=0
Settings
from libreyolo import LibreGen, SAMPLE_IMAGE model = LibreGen("direct3d-s2")result = model(SAMPLE_IMAGE, resolution=512)     # stops after the first passresult = model(SAMPLE_IMAGE, steps=(25, 15, 8))  # fewer steps per stage

A dense flow transformer finds the occupied 64^3 voxels, a sparse flow transformer with spatial sparse attention denoises a latent on them, and a sparse decoder turns it into a signed distance field at 512^3. A second pass repeats this at 1024^3. The mesh is the field's zero crossing, from marching cubes. Upstream runs on CUDA only (torchsparse, flash-attn, Triton kernels, a CUDA voxeliser); LibreYOLO replaces them with PyTorch operations and keeps the parameter names, so the released checkpoints load unchanged on CUDA, Apple MPS and CPU.

A seed gives the same object; on CUDA the starting noise equals upstream's for torch.manual_seed(seed). result.meshes holds vertices (V, 3), z-up in [-0.5, 0.5]^3, and faces (F, 3). Upstream's optional remeshing and hole filling are not included (they depend on GPL tools), so meshes come as marching cubes returns them.

It is slow in plain PyTorch: on an RTX A6000 the library measured about 18 minutes and 23 GB of GPU memory for one object at 1024^3, where upstream's CUDA kernels take 2.5 minutes. resolution=512 stops after the first pass.

Settings

SettingDefaultMeaning
resolution10241024 (upstream's default) or 512.
steps(50, 30, 15)Steps of the dense, 512 and 1024 stages; one integer sets all three.
guidance7.0Classifier-free guidance of all three stages.

Variants

One checkpoint, v11 (Direct3D-S2 v1.1): 2.0B parameters in eight networks, about 4.2 GB, plus 1.2 GB for DINOv2 ViT-L/14 and 0.4 GB for BiRefNet.

Export

ZadanieONNXTorchScriptExecuTorchTensorRTOpenVINOPaddleMNNRKNNAxelerancnnTFLiteCoreMLCore AI
3D generation3D generation to ONNX: brak obsługi3D generation to TorchScript: brak obsługi3D generation to ExecuTorch: brak obsługi3D generation to TensorRT: brak obsługi3D generation to OpenVINO: brak obsługi3D generation to Paddle: brak obsługi3D generation to MNN: brak obsługi3D generation to RKNN: brak obsługi3D generation to Axelera: brak obsługi3D generation to ncnn: brak obsługi3D generation to TFLite: brak obsługi3D generation to CoreML: brak obsługi3D generation to Core AI: brak obsługi

Export is not implemented for LibreGen 3D families. export(), train() and val() raise NotImplementedError (ADR 0031 in the library).

Licensing

Sprawdź licencję w repozytorium konkretnych pobieranych wag na Hugging Face. Każdy checkpoint w organizacji LibreYOLO ma licencję, która nie zawsze jest taka sama w całej rodzinie. To repozytorium jest źródłem rozstrzygającym, a poniższe podsumowanie opisuje stan z chwili ostatniej weryfikacji tej strony.

To opis obowiązujących licencji, a nie porada prawna. Jeśli odpowiedź ma znaczenie komercyjne, przeczytaj licencje i zasięgnij własnej porady prawnej.

Oryginalna praca
Direct3D-S2, DreamTech
Licencja projektu źródłowego
MIT
Kod źródłowy projektu
github.com/DreamTechAI/Direct3D-S2
Kod LibreYOLO
MIT
Wagi
MIT, ponownie opublikowane w huggingface.co/LibreYOLO
Interpretacja
MIT is a permissive license, so the code and weights can be used in commercial and closed-source products; the only obligation is to keep the copyright and license notice with copies. LibreYOLO's adapter is MIT; the flow transformers, SDF VAE decoders, refiners and spatial sparse attention it adapts from Direct3D-S2 keep MIT, with the attention following native-sparse-attention-triton (Apache-2.0) and the transformer modules TRELLIS's (MIT), noted in libreyolo/models/gen/direct3d_s2/NOTICE. Upstream's hole filling and remeshing, which need GPL-licensed pymeshfix and python-igraph, are not used. The wushuang98/Direct3D-S2 v1.1 weights (MIT) are redistributed unchanged as LibreYOLO/LibreDirect3DS2v11; the BiRefNet weights (MIT) come from LibreYOLO/LibreTRELLISl and the DINOv2 encoder, facebook/dinov2-with-registers-large (Apache-2.0), from Meta. All are pinned and none is gated.

Direct3D-S2 is by DreamTech (DreamTechAI/Direct3D-S2). Code and the wushuang98/Direct3D-S2 v1.1 weights are MIT, so commercial use is fine with the notice kept. LibreYOLO redistributes the weights unchanged as LibreYOLO/LibreDirect3DS2v11. BiRefNet (MIT) comes from LibreYOLO/LibreTRELLISl and the DINOv2 encoder, facebook/dinov2-with-registers-large (Apache-2.0), from Meta. Upstream's GPL pymeshfix and python-igraph are not used.

Citation

@article{wu2025direct3ds2gigascale3dgeneration,
  title={Direct3D-S2: Gigascale 3D Generation Made Easy with Spatial Sparse Attention}, 
  author={Shuang Wu and Youtian Lin and Feihu Zhang and Yifei Zeng and Yikang Yang and Yajie Bao and Jiachen Qian and Siyu Zhu and Philip Torr and Xun Cao and Yao Yao},
  journal={arXiv preprint arXiv:2505.17412},
  year={2025}
}

Skopiowano z bloku cytowania autorów w github.com/DreamTechAI/Direct3D-S2#-citation.

Zweryfikowano z LibreYOLO v1.8.0. Tabele obsługi, checkpointy i wyniki benchmarków na tej stronie są generowane na podstawie wydanej biblioteki i opublikowanych wag, a nie wpisywane ręcznie.