{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/direct3d-s2-gigascale-3d-generation-made-easy","title":"Direct3D-S2: Gigascale 3D Generation Made Easy with Spatial Sparse Attention","arxiv_id":"2505.17412","date":"2025-05-23","proceeding":null,"authors":["Shuang Wu","Youtian Lin","Feihu Zhang","Yifei Zeng","Yikang Yang","Yajie Bao","Jiachen Qian","Siyu Zhu","Xun Cao","Philip Torr","Yao Yao"],"abstract":"Generating high-resolution 3D shapes using volumetric representations such as Signed Distance Functions (SDFs) presents substantial computational and memory challenges. We introduce Direct3D-S2, a scalable 3D generation framework based on sparse volumes that achieves superior output quality with dramatically reduced training costs. Our key innovation is the Spatial Sparse Attention (SSA) mechanism, which greatly enhances the efficiency of Diffusion Transformer (DiT) computations on sparse volumetric data. SSA allows the model to effectively process large token sets within sparse volumes, substantially reducing computational overhead and achieving a 3.9x speedup in the forward pass and a 9.6x speedup in the backward pass. Our framework also includes a variational autoencoder (VAE) that maintains a consistent sparse volumetric format across input, latent, and output stages. Compared to previous methods with heterogeneous representations in 3D VAE, this unified design significantly improves training efficiency and stability. Our model is trained on public available datasets, and experiments demonstrate that Direct3D-S2 not only surpasses state-of-the-art methods in generation quality and efficiency, but also enables training at 1024 resolution using only 8 GPUs, a task typically requiring at least 32 GPUs for volumetric representations at 256 resolution, thus making gigascale 3D generation both practical and accessible. Project page: https://www.neural4d.com/research/direct3d-s2.","url_abs":"https://arxiv.org/abs/2505.17412v2","url_pdf":"https://arxiv.org/pdf/2505.17412v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"direct3d-s2-gigascale-3d-generation-made-easy","repo_url":"https://github.com/DreamTechAI/Direct3D-S2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"3d-generation","task_name":"3D Generation"},{"task_slug":"3d-object-reconstruction","task_name":"3D Object Reconstruction"},{"task_slug":"3d-reconstruction","task_name":"3D Reconstruction"},{"task_slug":"3d-shape-generation","task_name":"3D Shape Generation"},{"task_slug":"3d-shape-modeling","task_name":"3D Shape Modeling"},{"task_slug":"3d-geometry","task_name":"3D geometry"},{"task_slug":"image-to-3d","task_name":"Image to 3D"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2505.17412","atlas_url":"https://app.syntology.ai/?focus=2505.17412","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2505.17412"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/DreamTechAI/Direct3D-S2","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":1,"ran_draft_wrong":2,"unverified":7},"by_repo_kind":{"listed":{"samples":10,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"802333c05f6b9927","entry":"get_activation","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/models/refiner/unet3d.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/models/refiner/unet3d.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"802333c05f6b9927"}},{"code_sha256_prefix":"03310bba324ae4fb","entry":"modulate","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/modules/utils.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/modules/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"03310bba324ae4fb"}},{"code_sha256_prefix":"129b804760b3115f","entry":"zero_module","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/modules/utils.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/modules/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"129b804760b3115f"}},{"code_sha256_prefix":"484f913e371f7172","entry":"get_down_block","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/models/refiner/unet3d.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/models/refiner/unet3d.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"484f913e371f7172"}},{"code_sha256_prefix":"97a1118ec5c59ded","entry":"get_up_block","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/models/refiner/unet3d.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/models/refiner/unet3d.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"97a1118ec5c59ded"}},{"code_sha256_prefix":"6818f26f5c2a59a4","entry":"norm_layer","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/models/autoencoders/dense_vae.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/models/autoencoders/dense_vae.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6818f26f5c2a59a4"}},{"code_sha256_prefix":"e6b8cd9f318cc00b","entry":"patchify","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/modules/spatial.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/modules/spatial.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e6b8cd9f318cc00b"}},{"code_sha256_prefix":"9e5521078878a448","entry":"pixel_shuffle_3d","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/modules/spatial.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/modules/spatial.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9e5521078878a448"}},{"code_sha256_prefix":"9136b5f9b735334b","entry":"scale_module","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/modules/utils.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/modules/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9136b5f9b735334b"}},{"code_sha256_prefix":"d01c19a64a056d04","entry":"unpatchify","repo":"DreamTechAI/Direct3D-S2","repo_kind":"listed","path":"direct3d_s2/modules/spatial.py","file_url":"https://github.com/DreamTechAI/Direct3D-S2/blob/HEAD/direct3d_s2/modules/spatial.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d01c19a64a056d04"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}