Is visual image segmentation a bottom-up or an interactive process?

Abstract

Visualimage segmentation is the process by which the visual system groups features that are part of a single shape. Is image segmentation a bottom-up or an interactive process? In Experiments 1 and 2, we presented subjects with two overlapping shapes and asked them to determine whether two probed locations were on the same shape or on different shapes. The availability of top-down support was manipulated by presenting either upright or rotated letters. Subjects were fastest to respond when the shapes corresponded to familiar shapes-the upright letters. InExperiment 3, we used a variant of this segmentation task to rule out the possibility that subjects performed same/different judgments after segmentation and recognition of both letters. Finally,in Experiment 4, we ruled out the possibility that the advantage for upright letters was merely due to faster recognition of upright letters relative to rotated letters. The results suggested that the previous effects were not due to faster recognition of upright letters; stimulus familiarity influenced segmentation per se. The results are discussed in terms of an interactive model of visual image segmentation.Imagine a rider atop a horse. There are clearly points ofoverlap between the two objects, such as where the rider's legs occlude the side ofthe horse. For the visual system to recognize these two distinct objects, each visual object representation stored in visual memory must receive input from the regions of the visual field that correspond to that object and only to that object. Visual image segmentation is the process by which the visual system groups or "binds" locations and features that are part of a single shape. Is image segmentation a purely bottomup, stimulus-driven process or can stored knowledge of objects playa role in image segmentation?Both answers seem equally plausible. Bottom-up image segmentation is consistent with Fodor's (1983) claims of information encapsulation in the peripheral or input sys-

Visually-guided Attention Enhances Target Identification in a Complex Auditory Scene

“…In experiment 2, speech stimuli were used to address the question of whether long-term familiarity with sound sources influences the effectiveness of the different cues. In the visual literature, for example, there is evidence that object-based attention operates more effectively on familiar stimuli (Vecera and Farah 1997;Zemel et al 2002) and that the speed of visual search depends on target and distractor familiarity (Wang et al 1994;Shen and Reingold 2001).…”

Graded effects in hierarchical figure-ground organization: Reply to Peterson (1999).

“…As we have noted, most theoretical accounts of visual processing have placed figure-ground organization, as well as other modes of perceptual organization, prior to object identification in the visual processing stream (e.g., Biederman, 1987;Kosslyn, 1987;Mart, 1982;Neisser, 1967;Palmer & Rock, 1994a, 1994bVecera & Farah, 1997;Vecera & O'Reilly, 1998;Wallach, 1949). This placement is based on logical grounds: Presumably the retinal array must be organized before a perceiver can recognize the objects that appear in this array.…”

“…In our model, top-down feedback from object representations partially guides figure-ground processes (Vecera & O'Reilly, 1998; also see Vecera & Farah, 1997), and figure-ground processes are influenced simultaneously by two forces, bottom-up stimulus cues contained in the image and top-down, feedback cues from visual object representations stored in visual memory. Constraint satisfaction processes that emerge from interactive models allow both sets of cues to partially guide the network in determining what region is figure.…”

Vision Goes Symbolic Without Loss of Information Within the Preattentive Vision Phase: The Need to Shift the Learning Paradigm from Machine-Learning (from Examples) to Machine-Teaching (by Rules) at the First Stage of a Two-Stage Hybrid Remote... Part II

“…This means that in RS common practice any first-stage image segmentation algorithm is simultaneously affected by both omission and commission segmentation errors. Although the inherent ill-posedness of image segmentation is acknowledged by a reasonable portion of existing literature (Burr & Morrone, 1992;Corcoran et al, 2010;Corcoran & Winstanley, 2007;Delves et al, 1992;Hay & Castilla, 2006;Matsuyama & Shang-Shouq Hwang, 1990;Petrou & Sevilla, 2006;Vecera & Farah, 1997), this is often forgotten by a large segment of the RS community where literally dozens of "novel" segmentation algorithms are published each year (Zamperoni, 1996) (refer to Part I Section 2.4.1.2).  Semantic nets lack flexibility and scalability to cope with changes in sensor characteristics and users' changing needs, i.e., they are unsuitable for commercial RS image processing software toolboxes and remain limited to scientific applications.…”

“…In practical contexts this means the following.  Unlabeled (unsupervised) data learning algorithms, namely, unlabeled data clustering (Backer & Jain, 1981;Baraldi & Alpaydin, 2002a;Baraldi & Alpaydin, 2002b;Cherkassky & Mulier, 2006;Fritzke, 1997) and unlabeled (2-D) image segmentation algorithms (Burr & Morrone, 1992;Corcoran et al, 2010;Corcoran & Winstanley, 2007;Delves et al, 1992;Hay & Castilla, 2006;Matsuyama & ShangShouq Hwang, 1990;Petrou & Sevilla, 2006;Vecera & Farah, 1997), are recognized as inherently ill-posed problems subjective in nature by a relevant portion of existing literature.  Labeled (supervised) data learning classifiers are unable to establish correlation relationships between objective sensory (e.g., RS) data and categorical variables (e.g., land cover classes) at large data scale or fine semantic granularity.…”