Abstract
Inferring step-wise actions to assemble 3D objects with primitive bricks from images is a challenging task due to complex constraints and the vast number of possible combinations. Recent studies have demonstrated promising results on sequential LEGO brick assembly through the utilization of LEGO-Graph modeling to predict sequential actions. However, existing approaches are class-specific and require significant computational and 3D annotation resources. In this work, we first propose a computationally efficient breadth-first search (BFS) LEGO-Tree structure to model the sequential assembly actions by considering connections between consecutive layers. Based on the LEGO-Tree structure, we then design a class-agnostic tree-transformer framework to predict the sequential assembly actions from the input multi-view images. A major challenge of the sequential brick assembly task is that the step-wise action labels are costly and tedious to obtain in practice. We mitigate this problem by leveraging synthetic-to-real transfer learning. Specifically, our model is first pre-trained on synthetic data with full supervision from the available action labels. We then circumvent the requirement for action labels in the real data by proposing an action-to-silhouette projection that replaces action labels with input image silhouettes for self-supervision. Without any annotation on the real data, our model outperforms existing methods with 3D supervision by 7.8% and 11.3% in mIoU on the MNIST and ModelNet Construction datasets, respectively.
M. Guo, C. Li—Equal Contribution
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Arsalan Soltani, A., Huang, H., Wu, J., Kulkarni, T.D., Tenenbaum, J.B.: Synthesizing 3D shapes via modeling multi-view depth maps and silhouettes with deep generative networks. In: CVPR (2017)
Brochu, E., Cora, V.M., De Freitas, N.: A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. In: arXiv preprint arXiv:1012.2599 (2010)
Chang, A.X., et al.: Shapenet: an information-rich 3D model repository. In: arXiv preprint arXiv:1512.03012 (2015)
Chen, Y.C., Li, H., Turpin, D., Jacobson, A., Garg, A.: Neural shape mating: Self-supervised object assembly with adversarial shape priors. In: CVPR (2022)
Choy, C.B., Xu, D., Gwak, J., Chen, K., Savarese, S.: 3d-r2n2: a unified approach for single and multi-view 3D object reconstruction. In: ECCV (2016)
Chung, H., et al.: Brick-by-brick: combinatorial construction with deep reinforcement learning. In: NeurIPS (2021)
Curless, B., Levoy, M.: A volumetric method for building complex models from range images. In: SIGGRAPH (1996)
Debevec, P.E., Taylor, C.J., Malik, J.: Modeling and rendering architecture from photographs: a hybrid geometry-and image-based approach. In: SIGGRAPH (1996)
Dosovitskiy, A., et al.: An image is worth 16x16 words: transformers for image recognition at scale. ICLR (2021)
Eilers, S.: The lego counting problem. The American Mathematical Monthly (2016)
Fan, H., Su, H., Guibas, L.J.: A point set generation network for 3D object reconstruction from a single image. In: CVPR (2017)
Gadelha, M., et al.: Learning generative models of shape handles. In: CVPR (2020)
Gower, R., Heydtmann, A., Petersen, H.: Lego: Automated model construction (1998)
Jiang, L., Shi, S., Qi, X., Jia, J.: Gal: geometric adversarial loss for single-view 3d-object reconstruction. In: ECCV (2018)
Kar, A., Häne, C., Malik, J.: Learning a multi-view stereo machine. In: NeurIPS (2017)
Kim, J., Chung, H., Lee, J., Cho, M., Park, J.: Combinatorial 3D shape generation via sequential assembly. In: NeurIPS Workshop (2020)
Lee, S., Kim, J., Kim, J.W., Moon, B.R.: Finding an optimal lego® brick layout of voxelized 3D object using a genetic algorithm. In: The Genetic and Evolutionary Computation Conference (GECCO) (2015)
Lennon, K., et al.: Image2lego: Customized lego set generation from images. arXiv preprint arXiv:2108.08477 (2021)
Li, Y., Mo, K., Shao, L., Sung, M., Guibas, L.: Learning 3D part assembly from a single image. In: ECCV (2020)
Li, Y., Vinyals, O., Dyer, C., Pascanu, R., Battaglia, P.: Learning deep generative models of graphs. arXiv preprint arXiv:1803.03324 (2018)
Mandikal, P., Navaneet, K., Agarwal, M., Babu, R.V.: 3d-lmnet: latent embedding matching for accurate and diverse 3D point cloud reconstruction from a single image. In: BMVC (2018)
Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., Ng, R.: Nerf: representing scenes as neural radiance fields for view synthesis. In: ECCV (2020)
Park, J.J., Florence, P., Straub, J., Newcombe, R., Lovegrove, S.: Deepsdf: learning continuous signed distance functions for shape representation. In: CVPR (2019)
Paschalidou, D., Ulusoy, A.O., Geiger, A.: Superquadrics revisited: learning 3D shape parsing beyond cuboids. In: CVPR (2019)
Riegler, G., Osman Ulusoy, A., Geiger, A.: Octnet: learning deep 3D representations at high resolutions. In: CVPR (2017)
Schulman, J., Wolski, F., Dhariwal, P., Radford, A., Klimov, O.: Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347 (2017)
Shi, Z., Meng, Z., Xing, Y., Ma, Y., Wattenhofer, R.: 3d-retr: end-to-end single and multi-view 3D reconstruction with transformers. In: BMVC (2021)
Snavely, N., Seitz, S.M., Szeliski, R.: Photo tourism: exploring photo collections in 3D. In: SIGGRAPH (2006)
Sun, X., et al.: Pix3d: dataset and methods for single-image 3D shape modeling. In: CVPR (2018)
Tatarchenko, M., Dosovitskiy, A., Brox, T.: Octree generating networks: efficient convolutional architectures for high-resolution 3D outputs. In: ICCV (2017)
Thompson, R., Ghalebi, E., DeVries, T., Taylor, G.W.: Building lego using deep generative models of graphs. In: NeurIPS Workshop (2020)
Tulsiani, S., Efros, A.A., Malik, J.: Multi-view consistency as supervisory signal for learning shape and pose prediction. In: CVPR (2018)
Tulsiani, S., Gupta, S., Fouhey, D.F., Efros, A.A., Malik, J.: Factoring shape, pose, and layout from the 2D image of a 3D scene. In: CVPR (2018)
Tulsiani, S., Su, H., Guibas, L.J., Efros, A.A., Malik, J.: Learning shape abstractions by assembling volumetric primitives. In: CVPR (2017)
Vaswani, A., et al.: Attention is all you need. In: NeurIPS (2017)
Walsman, A., Zhang, M., Kotar, K., Desingh, K., Farhadi, A., Fox, D.: Break and make: interactive structural understanding using lego bricks. In: ECCV (2022)
Wang, R., Zhang, Y., Mao, J., Cheng, C.Y., Wu, J.: Translating a visual lego manual to a machine-executable plan. In: ECCV (2022)
Wu, J., Wang, Y., Xue, T., Sun, X., Freeman, B., Tenenbaum, J.: Marrnet: 3d shape reconstruction via 2.5 d sketches. In: NeurIPS (2017)
Wu, J., Zhang, C., Xue, T., Freeman, B., Tenenbaum, J.: Learning a probabilistic latent space of object shapes via 3d generative-adversarial modeling. In: NeurIPS (2016)
Wu, R., Tie, C., Du, Y., Zhao, Y., Dong, H.: Leveraging se (3) equivariance for learning 3d geometric shape assembly. In: ICCV (2023)
Wu, R., Zhuang, Y., Xu, K., Zhang, H., Chen, B.: Pq-net: a generative part seq2seq network for 3D shapes. In: CVPR (2020)
Wu, Z., et al.: 3D shapenets: a deep representation for volumetric shapes. In: CVPR (2015)
Xie, H., Yao, H., Sun, X., Zhou, S., Zhang, S.: Pix2vox: context-aware 3D reconstruction from single and multi-view images. In: ICCV (2019)
Xu, X., Guerrero, P., Fisher, M., Chaudhuri, S., Ritchie, D.: Unsupervised 3d shape reconstruction by part retrieval and assembly. In: CVPR (2023)
Zakka, K., Zeng, A., Lee, J., Song, S.: Form2fit: learning shape priors for generalizable assembly from disassembly. In: ICRA (2020)
Zhan, G., et al.: Generative 3d part assembly via dynamic graph learning. In: NeurIPS (2020)
Zou, C., Yumer, E., Yang, J., Ceylan, D., Hoiem, D.: 3d-prnn: generating shape primitives with recurrent neural networks. In: ICCV (2017)
Acknowledgement
This research/project is supported by the National Research Foundation Singapore and DSO National Laboratories under the AI Singapore Programme (Award Number: AISG2-RP-2020-016), and the Tier 2 grant MOE-T2EP20120-0011 from the Singapore Ministry of Education.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Supplementary material 2 (mp4 282 KB)
Rights and permissions
Copyright information
© 2025 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Guo, M., Li, C., Zhao, Y., Lee, G.H. (2025). TreeSBA: Tree-Transformer for Self-supervised Sequential Brick Assembly. In: Leonardis, A., Ricci, E., Roth, S., Russakovsky, O., Sattler, T., Varol, G. (eds) Computer Vision – ECCV 2024. ECCV 2024. Lecture Notes in Computer Science, vol 15144. Springer, Cham. https://doi.org/10.1007/978-3-031-73016-0_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-73016-0_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-73015-3
Online ISBN: 978-3-031-73016-0
eBook Packages: Computer ScienceComputer Science (R0)