w4 w gemini Flashcards
List the range of image properties to which V1 cells show selectivity.
colour
What is a hyper-column?
A hyper-column is a region of V1 that contains neurons covering the full range of RF types for a single spatial location.
Briefly describe the stimulus selectivity of simple cells in V1.
Simple cell: optimum response to an appropriately oriented stimulus
Briefly describe the stimulus selectivity of complex cells in V1.
Complex cell: optimum response to an appropriately oriented stimulus
Describe how simple cell responses could be modelled using convolution.
A simple cell RF can be well described by a Gabor function. Convolving the image with a Gabor mask will simulate the response of all simple cells selective for the same parameters across all hyper-columns. Repeating the convolution with Gabor masks with different parameters (e.g. orientation
Describe how complex cell responses could be modelled using convolution.
A complex cell can be modelled by combining the outputs of two or more simple cells. For example
Gabors functions are the components of natural images under the “sparsity” constraint. What is the sparsity constraint?
The sparsity constraint requires that the minimum number of components are present in each image.
How is the sparsity constraint relevant to efficient coding in the context of Gabor functions?
By using Gabors as the components by which an image is represented
Briefly describe what is meant by the classical receptive field (cRF).
Classical Receptive Field (cRF) = the region of visual space / the stimulus properties that can elicit a response from a neuron.
Briefly describe what is meant by the non-classical receptive field (ncRF).
Non-classical Receptive Field (ncRF) = the region of visual space / the stimulus properties that can modulate the response from a neuron
What is an “association field”? Describe the association field for a V1 cell with an orientation preference.
An association field is the pattern of long-range lateral connections received by an orientation selective neuron in V1. It defines the ncRF of such a neuron. A V1 neuron with a cRF selective for a particular orientation will receive lateral excitation from neighbouring V1 cells with similar orientation preferences that are aligned so that they are collinear or co-circular with it. It will receive lateral inhibition from other neighbouring V1 cells with similar orientation preferences.
How do lateral connections in V1 give rise to contour integration?
Contour integration is generated principally by lateral excitation between cells with nearly co-linear/co-circular orientation preferences. These cells enhance each others response
How do lateral connections in V1 give rise to pop-out?
Pop-out is generated principally by lateral inhibition between cells with similar preferences. These cells suppress each other’s response making cells responding to different image features relatively more active
How do lateral connections in V1 give rise to texture segmentation?
Texture segmentation is generated principally by lateral inhibition between cells with similar preferences. Hence
Briefly describe the difference between bottom-up and top-down influences on grouping.
Top-down influences come from prior knowledge and experience. They cause image elements to be grouped because of prior expectations about what elements belong to the same object. Bottom-up influences come from image properties. They cause image elements to be grouped because they have similar properties.
For image (a) (two rows of alternating black and white circles)
identify the Gestalt Law that gives rise to the observed grouping.
For image (b) (black circles grouped within ovals)
identify the Gestalt Law that gives rise to the observed grouping.
For image (c) (closely spaced black circles forming columns)
identify the Gestalt Law that gives rise to the observed grouping.
For image (d) (connected pairs of black circles)
identify the Gestalt Law that gives rise to the observed grouping.
For image (e) (incomplete squares)
identify the Gestalt Law that gives rise to the observed grouping.
For image (f) (a line of oriented bars)
identify the Gestalt Law that gives rise to the observed grouping.
Explain how lateral connections in V1 give rise to the Gestalt bias of similarity.
Lateral inhibitory connections cause mutual suppression of neurons representing similar image elements. At borders between dissimilar elements there is less inhibition
Explain how lateral connections in V1 give rise to the Gestalt bias of continuity.
Lateral excitatory connections cause mutual enhancement of neurons representing co-linearly orientated image elements. Hence
Explain what is meant by border ownership.
Border ownership refers to the fact that the boundary between two regions in an image is perceived as part of one region (the foreground) and not the other region (the background). This means that foreground objects have a defined shape (delineated by the border)
What is the role of V2 in border ownership?
V2 contains cells that encode border-ownership.
How could V2 cells compute border-ownership?
One mechanism by which V2 cells could compute border-ownership is via lateral connections within V2. Imagine that at each location there are multiple V2 neurons selective to different orientations. At each location and orientation there are a pair of neurons that prefer the foreground object to be on opposite sides of the border. These neurons compete with each other.
Describe the excitatory connections in a V2 border-ownership model.
Excitatory connections link neurons encoding segments consistent with a probable object.
Describe the inhibitory connections in a V2 border-ownership model.
Inhibitory connections link neurons encoding segments inconsistent with a probable object.
What is the sparsity constraint in the context of natural images?
The sparsity constraint requires that the minimum number of components are present in each image.
