Abstract
Many surface cues support three-dimensional shape perception, but humans can sometimes still see shape when these features are missing—such as when an object is covered with a draped cloth. Here we propose a framework for three-dimensional shape perception that explains perception in both typical and atypical cases as analysis-by-synthesis, or inference in a generative model of image formation. The model integrates intuitive physics to explain how shape can be inferred from the deformations it causes to other objects, as in cloth draping. Behavioural and computational studies comparing this account with several alternatives show that it best matches human observers (total n = 174) in both accuracy and response times, and is the only model that correlates significantly with human performance on difficult discriminations. We suggest that bottom-up deep neural network models are not fully adequate accounts of human shape perception, and point to how machine vision systems might achieve more human-like robustness.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$119.00 per year
only $9.92 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
Our behavioural data are publicly available at https://github.com/CNCLgithub/PbAS-model-human-comparisons. The experimental stimuli underlying the object-under-cloth task are publicly available at https://github.com/CNCLgithub/intuitive-physics-3d-shape-perception-stimuli.
Code availability
Code implementing the PbAS model, scripts for replicating model simulations, and a container for full reproducibility are publicly available at https://github.com/CNCLgithub/PbAS. Our custom Python scripts for data analysis are publicly available at https://github.com/CNCLgithub/PbAS-model-human-comparisons.
References
Bulthoff, H. Shape from X: psychophysics and computation. Comput. Models Vis. Process. 305–330 (1991).
Yildirim, I., Siegel, M. H. & Tenenbaum, J. B. Perceiving fully occluded objects via physical simulation. In Proc. 38th Annual Conference of the Cognitive Science Society 1265–1271 (Cognitive Science Society, 2016).
Phillips, F. & Fleming, R. W. The Veiled Virgin illustrates visual segmentation of shape by cause. Proc. Natl Acad. Sci. USA 117, 11735–11743 (2020).
Little, P. C. & Firestone, C. Physically implied surfaces. Psychol. Sci. 32, 799–808 (2021).
Wong, K. W., Bi, W., Soltani, A. A., Yildirim, I. & Scholl, B. J. Seeing soft materials draped over objects: a case study of intuitive physics in perception, attention, and memory. Psychol. Sci. 34, 111–119 (2022).
LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436–444 (2015).
Krizhevsky, A., Sutskever, I. & Hinton, G. E. Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25, 1097–1105 (2012).
Hong, H., Yamins, D. L., Majaj, N. J. & DiCarlo, J. J. Explicit information for category-orthogonal object properties increases along the ventral stream. Nat. Neurosci. 19, 613 (2016).
Yuille, A. & Kersten, D. Vision as Bayesian inference: analysis by synthesis? Trends Cogn. Sci. 10, 301–308 (2006).
Mumford, D. in Large-Scale Neuronal Theories of the Brain (eds Koch, C. & and Davis, J.) 125–152 (MIT Press, 1994).
Liu, Z., Knill, D. C. & Kersten, D. Object classification for human and ideal observers. Vis. Res. 35, 549–568 (1995).
Destler, N., Singh, M. & Feldman, J. Skeleton-based shape similarity. Psychol. Rev. https://doi.org/10.1037/rev0000412 (2023).
Erdogan, G. & Jacobs, R. A. Visual shape perception as Bayesian inference of 3D object-centered shape representations. Psychol. Rev. 124, 740 (2017).
Lee, M. J. & DiCarlo, J. J. An empirical assay of view-invariant object learning in humans and comparison with baseline image-computable models. Preprint at bioRxiv (2023).
Chandra, K., Li, T.-M., Tenenbaum, J. & Ragan-Kelley, J. Designing perceptual puzzles by differentiating probabilistic programs. In ACM SIGGRAPH 2022 Conference Proceedings 1–9 (ACM, 2022).
Chang, A. X. et al. ShapeNet: an information-rich 3D model repository. Preprint at https://doi.org/10.48550/arXiv.1512.03012 (2015).
Macklin, M., Müller, M., Chentanez, N. & Kim, T.-Y. Unified particle physics for real-time applications. ACM Trans. Graph. 33, 1–12 (2014).
Koch, E., Baig, F. & Zaidi, Q. Picture perception reveals mental geometry of 3D scene inferences. Proc. Natl Acad. Sci. USA 115, 7807–7812 (2018).
Snoek, J., Larochelle, H. & Adams, R. P. Practical Bayesian optimization of machine learning algorithms. Adv. Neural Inf. Process. Syst. 25, 2951–2959 (2012).
Cranmer, K., Brehmer, J. & Louppe, G. The frontier of simulation-based inference. Proc. Natl Acad. Sci. USA 117, 30055–30062 (2020).
Hamrick, J. B. & Griffiths, T. L. Mental rotation as Bayesian quadrature. In NIPS 2013 Workshop on Bayesian Optimization in Theory and Practice (2013).
Wang, A., Mei, S., Yuille, A. L. & Kortylewski, A. Neural view synthesis and matching for semi-supervised few-shot learning of 3D pose. Adv. Neural Inf. Process. Syst. 34, 7207–7219 (2021).
Järvenpää, M., Gutmann, M. U., Pleska, A., Vehtari, A. & Marttinen, P. Efficient acquisition rules for model-based approximate Bayesian computation. Bayesian Anal. 14, 595–622 (2019).
Kandasamy, K., Schneider, J. & Póczos, B. Bayesian active learning for posterior estimation. In 24th International Joint Conference on Artificial Intelligence 3605–3611 (PMLR, 2015).
Tamura, R. & Hukushima, K. Bayesian optimization for computationally extensive probability distributions. PLoS ONE 13, e0193785 (2018).
Schrimpf, M. et al. Brain-score: which artificial neural network for object recognition is most brain-like? Preprint at bioRxiv https://doi.org/10.1101/407007 (2018).
Yamins, D. L. & DiCarlo, J. J. Using goal-driven deep learning models to understand sensory cortex. Nat. Neurosci. 19, 356 (2016).
Deng, J. et al. Imagenet: a large-scale hierarchical image database. IEEE Conference on Computer Vision and Pattern Recognition 248–255 (IEEE, 2009).
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. IEEE Conference on Computer Vision and Pattern Recognition 770–778 (IEEE, 2016).
Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. Preprint at https://doi.org/10.48550/arXiv.1409.1556(2014).
Usher, M. & McClelland, J. L. The time course of perceptual choice: the leaky, competing accumulator model. Psychol. Rev. 108, 550 (2001).
Echeveste, R., Aitchison, L., Hennequin, G. & Lengyel, M. Cortical-like dynamics in recurrent circuits optimized for sampling-based probabilistic inference. Nat. Neurosci. 23, 1138–1149 (2020).
Yildirim, I., Belledonne, M., Freiwald, W. & Tenenbaum, J. Efficient inverse graphics in biological face processing. Sci. Adv. 6, eaax5979 (2020).
DiCarlo, J. J., Zoccolan, D. & Rust, N. C. How does the brain solve visual object recognition? Neuron 73, 415–434 (2012).
Yamins, D. L. Performance-optimized hierarchical models predict neural responses in higher visual cortex. Proc. Natl Acad. Sci. USA 111, 8619–8624 (2014).
Chen, T., Kornblith, S., Norouzi, M. & Hinton, G. A simple framework for contrastive learning of visual representations. In International Conference on Machine Learning 1597–1607 (PMLR, 2020).
Radford, A. et al. Learning transferable visual models from natural language supervision. In International Conference on Machine Learning 8748–8763 (PMLR, 2021).
Dosovitskiy, A. et al. An image is worth 16x16 words: transformers for image recognition at scale. Preprint at https://doi.org/10.48550/arXiv.2010.11929 (2020).
Konkle, T. & Alvarez, G. A. A self-supervised domain-general learning framework for human ventral stream representation. Nat. Commun. 13, 1–12 (2022).
Geirhos, R. Partial success in closing the gap between human and machine vision. Adv. Neural Inf. Process. Syst. 34, 23885–23899 (2021).
Gilbert, C. D. In Principles of Neural Science 5th edn (eds. Kandel, E. R. et al.) 556–576 (McGraw-Hill, 2013).
Baker, N., Lu, H., Erlikhman, G. & Kellman, P. J. Deep convolutional networks do not classify based on global object shape. PLoS Comput. Biol. 14, e1006613 (2018).
Feldman, J. & Singh, M. Information along contours and object boundaries. Psychol. Rev. 112, 243 (2005).
Zucker, S. W. On qualitative shape inferences: a journey from geometry to topology. Preprint at https://doi.org/10.48550/arXiv.2008.08622 (2020).
Kunsberg, B. & Zucker, S. W. Critical contours: an invariant linking image flow with salient surface organization. SIAM J. Imaging Sci. 11, 1849–1877 (2018).
Grill-Spector, K. & Kanwisher, N. Visual recognition: as soon as you know it is there, you know what it is. Psychol. Sci. 16, 152–160 (2005).
Ullman, S. in Readings in Computer Vision (eds Fischler, M. A. & Firschein, O.) 298–328 (Elsevier, 1987).
Shepard, R. N. & Metzler, J. Mental rotation of three-dimensional objects. Science 171, 701–703 (1971).
Shams, L. & Beierholm, U. Bayesian causal inference: a unifying neuroscience theory. Neurosci. Biobehav. Rev. 137, 104619 (2022).
Fischer, J., Mikhael, J. G., Tenenbaum, J. B. & Kanwisher, N. Functional neuroanatomy of intuitive physical inference. Proc. Natl Acad. Sci. USA 113, E5072–E5081 (2016).
Nash, C., Ganin, Y., Eslami, S. A. & Battaglia, P. Polygen: an autoregressive generative model of 3d meshes. In International Conference on Machine Learning (7220–7229) (2020).
Pfaff, T., Fortunato, M., Sanchez-Gonzalez, A. & Battaglia, P. W. Learning mesh-based simulation with graph networks. Preprint at https://doi.org/10.48550/arXiv.2010.03409 (2021).
Mrowca, D. et al. Flexible neural representation for physics prediction. In Proc. 32nd International Conference on Information Processing Systems 8813–8824 (2018).
Smith, K. et al. Modeling expectation violation in intuitive physics with coarse probabilistic object representations. Adv. Neural Inf. Process. Syst. 32, 8983–8993 (2019).
Piloto, L. S., Weinstein, A., Battaglia, P. & Botvinick, M. Intuitive physics learning in a deep-learning model inspired by developmental psychology. Nat. Hum. Behav. 6, 1257–1267 (2022).
Sanborn, A. N., Mansinghka, V. K. & Griffiths, T. L. Reconciling intuitive physics and Newtonian mechanics for colliding objects. Psychol. Rev. 120, 411 (2013).
Wu, J., Yildirim, I., Lim, J. J., Freeman, B. & Tenenbaum, J. Galileo: perceiving physical object properties by integrating a physics engine with deep learning. Adv. Neural Inf. Process. Syst. 28, 127–135 (2015).
Schwettmann, S., Tenenbaum, J. B. & Kanwisher, N. Invariant representations of mass in the human brain. eLife 8, e46619 (2019).
Bi, W., Shah, A. D., Wong, K. W., Scholl, B. & Yildirim, I. Perception of soft materials relies on physics-based object representations: Behavioral and computational evidence. Preprint at bioRxiv https://doi.org/10.1101/2021.05.12.443806 (2021).
Paulun, V. C., Schmidt, F., van Assen, J. J. R. & Fleming, R. W. Shape, motion, and optical cues to stiffness of elastic objects. J. Vis. 17, 20–20 (2017).
Paulun, V. C. & Fleming, R. W. Visually inferring elasticity from the motion trajectory of bouncing cubes. J. Vis. 20, 6–6 (2020).
Bates, C. J., Yildirim, I., Tenenbaum, J. B. & Battaglia, P. Modeling human intuitions about liquid flow with particle-based simulation. PLoS Comput. Biol. 15, e1007210 (2019).
Kubricht, J., Zhu, Y., Jiang, C., Terzopoulos, D., Zhu, S. C. & Lu, H. Consistent probabilistic simulation underlying human judgment in substance dynamics. In Proc. 39th Annual Conference of the Cognitive Science Society 3426–3431 (Cognitive Science Society, 2017).
Van Assen, J. J. R., Barla, P. & Fleming, R. W. Visual features in the perception of liquids. Curr. Biol. 28, 452–458 (2018).
Chen, Y.-C. & Scholl, B. J. The perception of history: seeing causal history in static shapes induces illusory motion perception. Psychol. Sci. 27, 923–930 (2016).
Fleming, R. W. & Schmidt, F. Getting “fumpered”: classifying objects by what has been done to them. J. Vis. 19, 15–15 (2019).
Schmidt, F., Phillips, F. & Fleming, R. W. Visual perception of shape-transforming processes: ‘shape scission’. Cognition 189, 167–180 (2019).
Blender Online Community Blender—a 3D modelling and rendering package. Blender Institute http://www.blender.org (2015).
Geirhos, R. et al. ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. Preprint at https://doi.org/10.48550/arXiv.1811.12231 (2018).
Rasmussen, C. E. & Williams, C. K. Gaussian Processes for Machine Learning (MIT Press, 2006).
Nogueira, F. Bayesian Optimization: open source constrained global optimization tool for Python. GitHub https://github.com/fmfn/BayesianOptimization (2014).
GPy: a Gaussian process framework in Python. GitHub http://github.com/SheffieldML/GPy (2012).
Schultz, M. & Joachims, T. Learning a distance metric from relative comparisons. Adv. Neural Inf. Process. Syst. 16, 41–48 (2003).
Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. Preprint at https://doi.org/10.48550/arXiv.1412.6980 (2015).
Reddi, S. J., Kale, S. & Kumar, S. On the convergence of Adam and beyond. Preprint at https://doi.org/10.48550/arXiv.1904.09237 (2018).
Acknowledgements
This work was supported by the Center for Brains, Minds and Machines (CBMM), funded by NSF STC award CCF-1231216; ONR MURI N00014-13-1-0333 (to J.B.T.); a grant from Toyota Research Institute (to J.B.T.); and a grant from Mitsubishi MELCO (to J.B.T.). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. A high performance computing cluster (OpenMind) was provided by the McGovern Institute for Brain Research. We thank K. Smith, B. Egger, K. Allen, G. Erdogan, M. Tenenbaum, N. Kanwisher and V. Paulun for their comments on a previous version of this manuscript.
Author information
Authors and Affiliations
Contributions
I.Y., M.H.S. and J.B.T. conceived and designed the study. I.Y. analysed the data. I.Y., M.H.S., A.A.S. and J.B.T. designed stimuli and experiments, and wrote and edited the manuscript. All authors contributed to the models.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Human Behaviour thanks Yaniv Morgenstern, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Supplementary Information
Supplementary Figs. 1–9.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Yildirim, I., Siegel, M.H., Soltani, A.A. et al. Perception of 3D shape integrates intuitive physics and analysis-by-synthesis. Nat Hum Behav 8, 320–335 (2024). https://doi.org/10.1038/s41562-023-01759-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41562-023-01759-7