z-logo
open-access-imgOpen Access
Seeing Global Motion in a Random Dot Image Sequence
Author(s) -
ChienChung Chen,
Hiroshi Ashida,
Xirui Yang,
PeiYin Chen
Publication year - 2020
Publication title -
i-perception
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.64
H-Index - 26
ISSN - 2041-6695
DOI - 10.1177/2041669520961104
Subject(s) - computer vision , mathematics , artificial intelligence , motion field , structure from motion , motion (physics) , motion estimation , computer science
In a stimulus with multiple moving elements, an observer may perceive that the whole stimulus moves in unison if (a) one can associate an element in one frame with one in the next (correspondence) and (b) a sufficient proportion of correspondences signal a similar motion direction (coherence). We tested the necessity of these two conditions by asking the participants to rate the perceived intensity of linear, concentric, and radial motions for three types of stimuli: (a) random walk motion, in which the direction of each dot was randomly determined for each frame, (b) random image sequence, which was a set of uncorrelated random dot images presented in sequence, and (c) global motion, in which 35% of dots moved coherently. The participants perceived global motion not only in the global motion conditions but also in the random image sequences, though not in random walk motion. The type of perceived motion in the random image sequences depends on the spatial context of the stimuli. Thus, although there is neither a fixed correspondence across different frames nor a coherent motion direction, observers can still perceive global motion in the random image sequence. This result cannot be explained by motion energy or local aperture border effects.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here
Accelerating Research

Address

John Eccles House
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom