Purchase this article with an account.
Moqian Tian, Daniel Yamins, Kalanit Grill-Spector; Learning the 3D structure of objects from 2D views depends on shape, not format. Journal of Vision 2016;16(12):410. doi: https://doi.org/10.1167/16.12.410.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Humans can learn to recognize new objects just from observing example views. However, it is unknown what structural information enables this learning. To address this question we manipulated the amount of structural information given to subjects during unsupervised learning by varying the format of the trained views. We then tested how format affected participants' ability to discriminate similar objects across views that were rotated 90° apart. We found that after training, participants' performance increased and generalized to new views in the same format. Surprisingly, the improvement was similar across line drawings, shape-from-shading, and shape-from-shading + stereo, even though the latter two formats provide richer depth information compared to line drawings. In contrast, participants' improvement was significantly lower when training used silhouettes, suggesting that silhouettes do not have enough information to generate a robust 3D structure. To test whether the learned object representations were format-specific or format-invariant, we examined if learning novel objects from example views transfers across formats. We found that learning objects from example line drawings transferred to shape-from-shading and vice versa. These results have important implications for theories of object recognition because they suggest that (1) learning the 3D structure of objects does not require rich structural cues during training as long as shape information of internal and external features is provided and (2) learning generates shape-based object representations independent of the training format.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only