Perceptual straightening of natural videos

109 citations recorded by the source corpus.

How researchers see this paper

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

Data-Efficient Image Recognition with Contrastive Predictive Coding (2019, background): “Indeed, human perceptual representations have been shown to linearize (or ‘straighten’) the temporal transformations found in natural videos, a property lacking from current supervised image recognition models (Hénaff et al., 2019), and theories of both spatial and temporal predictability have succe”
Interpreting neural computations by examining intrinsic and embedding dimensionality of neural activity (2021, background): “The intrinsic dimensionality for more natural stimuli has not yet been systematically quantified, but it is likely that it would increase with the complexity of spatial [28] and/or temporal [29,30] dependencies.”
Artificial Neural Networks for Neuroscientists: A Primer (2020, background): “While neural representations of natural images are overall similar between deep convolutional networks and mammalian visual systems, natural videos are represented by much smoother trajectories in humans compared to mainstream convolutional networks [Hénaff et al., 2019].”