Top-down and bottom-up processes for object segmentation

Object segmentation by top-down processes

In cluttered scenes, some object boundaries may not be marked by image cues. In such cases, the boundaries must be defined top-down as a result of object recognition. Here we ask if observers can retain the boundaries of several recognized objects in order to segment an unfamiliar object. We generated scenes consisting of neatly stacked objects, and the objects themselves consisted of neatly st...

Bottom-up, top-down

Unsupervised Tattoo Segmentation Combining Bottom-Up and Top-Down Cues

Tattoo segmentation is challenging due to the complexity and large variance in tattoo structures. We have developed a segmentation algorithm for finding tattoos in an image. Our basic idea is split-merge: split each tattoo image into clusters through a bottom-up process, learn to merge the clusters containing skin and then distinguish tattoo from the other skin via top-down prior in the image i...

Top-down and bottom-up processes in speech comprehension.

Speech comprehension includes both bottom-up and top-down processes, and imaging studies have isolated a frontal-temporal network of brain areas active during speech perception. However, the precise role of the various areas in this network during normal speech comprehension is not yet fully understood. In the present fMRI study, the signal-to-noise ratio (SNR) of spoken sentences was varied in...

Bottom-Up Beats Top-Down

In PEPA, the calculation of the transitions enabled by a process accounts for a large part of the time for the state space exploration of the underlying Markov chain. Unlike other approaches based on recursion, we present a new technique that is iterative—it traverses the process’ binary tree from the sequential components at the leaves up to the root. Empirical results shows that this algorith...