{"url":"/dataset/sba","name":"SBA","full_name":"Sequentail Brick Assembly Dataset","description_markdown":"The RAD (Randomly Assembled Object Construction) dataset is a synthetic 3D LEGO dataset designed for the task of Sequential Brick Assembly (SBA). Here are the key characteristics and details:\r\n\r\nHigh-level explanation:\r\n\r\nConsists of 3D objects built with 2x4 LEGO bricks. \r\nObjects are randomly assembled following predefined connection configurations.\r\nIncludes variations like RAD-S and RAD with different numbers of bricks per object.\r\nProvides ground truth labels for assembly actions.\r\nIncludes multi-view images, sequential assembly actions, 3D voxel, conjunction graph, LEGO ldr file, etc.\r\n\r\nMotivations and content summary:\r\n\r\nCreated to provide a large-scale synthetic dataset for training and evaluating SBA models.\r\nAllows for controlled experimentation with varying object complexity.\r\nRAD-1k: 1000 objects with around 15 bricks each.\r\nRAD-S: 10000 objects built with around 20 bricks each.\r\nRAD: 10000 objects built with around 60 bricks each.\r\nExpands on previous datasets by allowing more flexible brick connections, including bottom-to-top assembly.\r\n\r\nPotential use cases:\r\n\r\nPre-training models for sequential brick assembly tasks.\r\nEvaluating the scalability of assembly algorithms with increasing object complexity.\r\nStudying transfer learning from synthetic to real-world assembly tasks.\r\nBenchmarking performance of different SBA approaches.\r\nInvestigating generalization capabilities of assembly models to different object structures.\r\n\r\nThe RAD dataset provides a valuable resource for researchers working on 3D object assembly, particularly in developing and testing algorithms that can handle increasingly complex structures built from simple primitives.","description_withheld":null,"homepage":"https://dreamguo.github.io/projects/TreeSBA/","introduced_date":"2024-07-22","introduced_date_note":null,"introduced_by":{"paper":"/paper/treesba-tree-transformer-for-self-supervised","title":"TreeSBA: Tree-Transformer for Self-Supervised Sequential Brick Assembly","first_author":"Mengqi Guo","url":null},"license":{"name":"MIT","url":null},"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"3D","url":"/datasets/modality/3d"},{"name":"3d meshes","url":"/datasets/modality/3d-meshes"},{"name":"Actions","url":"/datasets/modality/actions"}],"tasks":[{"name":"3D Assembly","url":"/task/3d-assembly","datasets_with_task":"/datasets/task/3d-assembly"}],"languages":[],"variants":["SBA"],"data_loaders":[],"num_papers_in_archive":1,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[],"papers_with_a_benchmark_row":[],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}