{"id":30221,"date":"2013-10-14T12:13:51","date_gmt":"2013-10-14T12:13:51","guid":{"rendered":"http:\/\/www.labri.fr\/perso\/barla\/blog\/?p=30221"},"modified":"2014-02-12T17:52:52","modified_gmt":"2014-02-12T17:52:52","slug":"prism2","status":"publish","type":"post","link":"https:\/\/www.labri.fr\/perso\/barla\/blog\/?p=30221","title":{"rendered":"Roland @ PRISM2"},"content":{"rendered":"<p id=\"top\" \/>\n<blockquote><p>It seems to me that portions of the orientation fields you&#8217;ve shown in your talk are more reliable to infer shape than others.<br \/>\nI had the impression that these portions correspond to places where the field divergence is low, which seems to occur when curvature is high in one direction.<br \/>\nI wonder whether you&#8217;ve tried measuring _curl and divergence of your flows. If not, we could probably try that with one of our Patates that does that<\/p><\/blockquote>\n<p>We are doing a series of experiments right now on which regions are most informative about different shape properties. We haven&#8217;t looked at div, but _curl is something we have been thinking about. The normal fields of real surfaces must have zero _curl, and we are interested in asking what subjects see when we show them impossible stimuli (a bit like the Escher staircase). Do you have a specific geometrical intuition for why low divergence should be good? High divergence correlates with low tangential curvature and high normal curvature. I still haven&#8217;t got a very strong intuition about what this means in terms of the surface, although Ben Kunsberg has worked this all out mathematically. I think low normal curvature may be a good thing as it tends to mean ridge structure, although there are situations near cusps where the shape is really visible and normal curvature gets large. So it is not as simple as it looks at first glance.<\/p>\n<blockquote><p>It also seemed to me that the weird shading is more sensitive to curvature, so that curvature magnitude is more prominent, with direction and sign unchanged. Thinking about this in terms of BRDF, it would make sense: with glossy materials, BRDF lobes not only depend on normals, but also exaggerate their variations. I wonder whether this difference in magnitude between diffuse and glossy\/specular shading has ever been tested experimentally.<\/p><\/blockquote>\n<p>Absolutely &#8230; a great observation, and I agree. Mirrors also look more &#8216;curvy&#8217; than shaded version of the same object. We still don&#8217;t really have a good psychophysical task for measuring perceived magnitude of curvature (or indeed any second order properties of surfaces), but it would probably be easy to make a cartoon task involving cylinder-like or ellipsoidal objects where subjects adjust a single curvature parameter to report what they see. It would be fun to mix this with studies on anisotropy and how this effects perceived shape.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>It seems to me that portions of the orientation fields you&#8217;ve shown in your talk are more reliable to infer shape than others. I had the impression that these portions correspond to places where the field divergence is low, which seems to occur when curvature is high in one direction. I wonder whether you&#8217;ve tried &#8230; <a title=\"Roland @ PRISM2\" class=\"read-more\" href=\"https:\/\/www.labri.fr\/perso\/barla\/blog\/?p=30221\" aria-label=\"Read more about Roland @ PRISM2\">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":[21],"tags":[],"class_list":["post-30221","post","type-post","status-publish","format-standard","hentry","category-discuss"],"_links":{"self":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/30221","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=30221"}],"version-history":[{"count":14,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/30221\/revisions"}],"predecessor-version":[{"id":34271,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/30221\/revisions\/34271"}],"wp:attachment":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=30221"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=30221"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=30221"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}