The Invariance Hypothesis Implies Domain-Specific Regions in Visual Cortex

By Joel Z. Leibo, Qianli Liao, Fabio Anselmi, Tomaso A. Poggio. 2014

35 citations recorded by the source corpus.

Is visual cortex made up of general-purpose information processing machinery, or does it consist of a collection of specialized modules? If prior knowledge, acquired from learning a set of objects is only transferable to new objects that share properties with the old, then the recognition system’s optimal organization must be one containing specialized modules for different object classes. Our analysis starts from a premise we call the invariance hypothesis: that the computational goal of the ventral stream is to compute an invariant-to-transformations and discriminative signature for recognition. The key condition enabling approximate transfer of invariance without sacrificing discriminability turns out to be that the learned and novel objects transform similarly. This implies that the optimal recognition system must contain subsystems trained only with data from similarly-transforming objects and suggests a novel interpretation of domain-specific regions like the fusiform face area (FFA). Furthermore, we can define an index of transformation-compatibility, computable from videos, that can be combined with information about the statistics of natural vision to yield predictions for which object categories ought to have domain-specific regions in agreement with the available data. The result is a unifying account linking the large literature on view-based recognition with the wealth of experimental evidence concerning domain-specific regions.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

View-tolerant face recognition and Hebbian learning imply mirror-symmetric neural tuning to head orientation (2016, positive): “Several hierarchical models of object recognition [17, 40, 43, 9] and face recognition [8, 29, 14] feature a progression from view-specific early processing stages to view-invariant later processing stages similar to ML/MF and AM, respectively.”
On Invariance and Selectivity in Representation Learning (2015, methodology): “Preliminary answers to these questions are described in [3,5,27,30].”
Intelligent Information Loss: The Coding of Facial Identity, Head Pose, and Non-Face Information in the Macaque Face Patch System (2015, methodology): “…across posterior, middle, and anterior patch, it should be possible to gain a better understanding of the visual features that drive the populations of neurons recorded from in this study, and potentially relate these features to computational models (Leibo et al., 2014; Tan and Poggio, 2014).”