{"id":40041,"date":"2019-08-01T06:33:08","date_gmt":"2019-08-01T06:33:08","guid":{"rendered":"http:\/\/www.labri.fr\/perso\/barla\/blog\/?p=40041"},"modified":"2019-08-01T06:33:09","modified_gmt":"2019-08-01T06:33:09","slug":"shape-as-an-exploratory-process","status":"publish","type":"post","link":"https:\/\/www.labri.fr\/perso\/barla\/blog\/?p=40041","title":{"rendered":"Shape as an exploratory process"},"content":{"rendered":"<p id=\"top\" \/>\n<h3 class=\"wp-block-heading\">Contours first<\/h3>\n\n\n\n<p>Individual &#8220;aperture-sized&#8221; observations do not permit to elicit an impression of shape; like the many &#8220;leg-persona&#8221; of the sea urchin that sense the environment but never come together to yield a holistic shape percept at the scale of the organism [von Uexk\u00fcll].<\/p>\n\n\n\n<p>Look at a common object, like a stone, but only in small chunks: it makes it hard to grasp the overall shape, especially when the contours are not found in the aperture. Let the apertures grow until they encompass at least part of the object contours; then shape as a coherent percept emerges. <\/p>\n\n\n\n<p>This suggests that shape perception works from the outside in: it first grasps the contours of the object under scrutiny; this leaves several patterns that need to be made sense of <em>in the context<\/em> of these contours. The exploratory process is thus initiated and maintained by the observer until another task is &#8220;launched&#8221; or boredom sets in&#8230;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">A matter of scales<\/h3>\n\n\n\n<p>The scale at which the object shape is visually probed largely depends on the observer: his position, current task, attention, etc. Widely different organisms will thus apprehend a same object as different entities. One may argue that there is thus no objective definition of an object (!), or more generally of the world: each organism has its own Umwelt (von Uexk\u00fcll again). Yet, at a given scale (of one or more organisms), an exhaustive object shape definition might be given, with details not available to perception conveniently folded in material properties as done in Computer Graphics.<\/p>\n\n\n\n<p>This is what is usually referred as a <em>physical description, <\/em>even though here it is made explicit that it depends on scale: we might talk of the shape of a mountain, or the shape of  a grain of sand. In all cases, shape might be described in mathematical (i.e., geometrical) terms. This provides a convenient &#8220;ground truth&#8221; (albeit constructed) to study which aspects of object shape are available to perception and how.<\/p>\n\n\n\n<p>Mathematical descriptions of surface shape involve local (infinitesimal) properties such as orientations (normals) and curvatures (convexities, concavities, saddle points). A more global description requires to establish connections among neighbor surface points.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Subjective shape<\/h3>\n\n\n\n<p>Once a viewpoint is chosen, object shape turns into an ever more subjective: because of foreshortening, some surface regions are compressed, others are hidden from view and <em>occluding contours<\/em> emerge. These are fundamental as they do not depend on the material the object is made of; hence they are generically present irrespective of lighting. This might be the reason of their prominent role: occluding contours are nearly always <em>potentially available<\/em> cues to object shape; they only need a slight motion of the object or view to becomes <em>effectively available<\/em>. <\/p>\n\n\n\n<p>Other cues become potentially available once one considers <em>material and lighting<\/em>: a variation in pigmentation is affected by surface orientation; a glossy reflection is distorted by surface curvature. However, these cues require a proper lighting to be effectively available. Hence a natural strategy to better perceive shape is to probe objects through relative motions of either the viewpoint, object or lighting, which turns potential into effective cues. The way these are correlated and connected to occluding contours remains a central challenge of ecological optics. Their behavior in motion may also be studied from a geometric standpoint at the current scale of interest.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">The exploratory process<\/h3>\n\n\n\n<p>We&#8217;re now ready to get back to the exploratory process of shape perception. Take the example of a simple stone: one may grab it in his hands by identification of its occluding contours. Then, paying attention to its overall shape, the observer may <em>pick-up<\/em> effectively available cues such as the way shadows and reflections unfold, which are diagnostic of the shape of the stone and naturally comply with its contours. However, some regions might lack in contrast or details, hence the observer might <em>probe<\/em> other aspects of shape by turning the stone in his hand, turning potential cues into effective ones that can be readily picked up: for instance, the way reflections flow toward the probed regions, compress and stretch, merge and separate.<\/p>\n\n\n\n<p>However, by doing so, the observer might pick up new visual patterns that cannot be explained by a smoothly curving stone shape, Now, by adjusting the scale of interest down to smaller details, these patterns quickly turn out to be due to tiny relief crevices, which themselves provide effective cues through reflective (even translucent) patterns. Now, by probing the object with a grazing light direction, these tiny crevices appear to be nearly all over the stone; they were merely unattended during the previous exploration steps.<\/p>\n\n\n\n<p>At some point the observer throws the stone away, which makes it land on a pile. Now the scale of interest is that of the pile; and even though individual stones can quickly be attended to by a swift change of scale, the overall shape of the pile can be probed as was done previously, provided enough stones are accumulated to create relevant visual patterns.<\/p>\n\n\n\n<p>At this scale, the contours, surface orientations and curvatures of the pile may be seen to emerge from the enveloppe of the stones. But this is only a partial surface in the mathematical sense, as it becomes fragmented when a stone is missing and leaves a hole for instance. This may be either ignored, or used as a trigger to switch scale again to focus on individual stones. <\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Scale jumps and indirections<\/h3>\n\n\n\n<p>There is not usually a continuous change of scales: for instance, going from a leaf, to a tree, to a forest canopy, to a mountain involves jumps in scale. What, then, makes a scale stable? It might well be its propension to produce potential shape cues (even partial ones); hence relevant scales might be revealed by probing the environment, so that there are always more to discover.<\/p>\n\n\n\n<p>Some of these relevant scales may even be attended to through other objects, simply because they produce a smooth image of the environment: smoothly curving reflective or transparent surfaces mostly. The quest for apparent shape seems to be boundless!<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Contours first Individual &#8220;aperture-sized&#8221; observations do not permit to elicit an impression of shape; like the many &#8220;leg-persona&#8221; of the sea urchin that sense the environment but never come together to yield a holistic shape percept at the scale of the organism [von Uexk\u00fcll]. Look at a common object, like a stone, but only in &#8230; <a title=\"Shape as an exploratory process\" class=\"read-more\" href=\"https:\/\/www.labri.fr\/perso\/barla\/blog\/?p=40041\" aria-label=\"Read more about Shape as an exploratory process\">Read more<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[651],"tags":[],"class_list":["post-40041","post","type-post","status-publish","format-standard","hentry","category-thoughts"],"_links":{"self":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/40041","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=40041"}],"version-history":[{"count":7,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/40041\/revisions"}],"predecessor-version":[{"id":40048,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/40041\/revisions\/40048"}],"wp:attachment":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=40041"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=40041"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=40041"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}