evutils.vis.reconstructor.metavision#

Event-to-video reconstruction using the Metavision ML model.

Attributes#

torch

F

Classes#

Metavision_Reconstructor

Reconstructor using the Metavision model.

Module Contents#

evutils.vis.reconstructor.metavision.torch = False#
evutils.vis.reconstructor.metavision.F#
class evutils.vis.reconstructor.metavision.Metavision_Reconstructor(width: int, height: int, args: dict | None = None)#

Bases: evutils.vis.reconstructor._base.Reconstructor

Inheritance diagram of evutils.vis.reconstructor.metavision.Metavision_Reconstructor

Reconstructor using the Metavision model.

Parameters:

  • width (int) – Width of the frame

  • height (int) – Height of the frame

  • args (dict, optional) – Additional arguments for the Metavision reconstructor, by default {}

model#
gen_frame(e: numpy.ndarray) → numpy.ndarray#

Reconstructs a single frame from the given events.

Parameters:

e (np.ndarray) – Array of events.

Returns:

Reconstructed grayscale image frame as a numpy array of shape (height, width) with dtype uint8.

Return type:

np.ndarray