{"id":23531,"date":"2012-08-29T08:49:07","date_gmt":"2012-08-29T08:49:07","guid":{"rendered":"http:\/\/www.labri.fr\/perso\/barla\/blog\/?p=23531"},"modified":"2012-08-29T08:49:07","modified_gmt":"2012-08-29T08:49:07","slug":"siggraph-2012","status":"publish","type":"post","link":"https:\/\/www.labri.fr\/perso\/barla\/blog\/?p=23531","title":{"rendered":"Siggraph 2012"},"content":{"rendered":"<p id=\"top\" \/>\n<h4>Schelling Points<\/h4>\n<ul>\n<li> They claim that other descriptors are poorly related to theirs<\/li>\n<li>Same with GLS?<\/li>\n<\/ul>\n<h4>Functional maps<\/h4>\n<ul>\n<li> See Conformal flattening [Lipman]<\/li>\n<li>Use Laplace-Beltrami eigen values as functions<\/li>\n<li>Try with GLS?<\/li>\n<\/ul>\n<h4>Sketch-based retrieval<\/h4>\n<ul>\n<li> Use Gabor filters to characterize line-based renderings of 3D objects from multiple views<\/li>\n<li>Use quantization to &#8220;reduce&#8221; to a 1000 dim-descriptor<\/li>\n<li>Try replacing Gabor with ILA gradient fields?<\/li>\n<\/ul>\n<h4>Adaptive manifolds<\/h4>\n<ul>\n<li> Find manifolds in the n-dimensional representation of an image<\/li>\n<li>Permits to drastically reduce the number of evaluated samples<\/li>\n<\/ul>\n<h4>Monte-Carlo filtering<\/h4>\n<ul>\n<li> Use mutual information to detect which variations are due to scene features, and which ones are due to random parameters.<\/li>\n<li>Why not instead study 1st order variations of the rendering equation with lenses and sampling patterns?<\/li>\n<\/ul>\n<h4>Global Illumination reconstruction<\/h4>\n<ul>\n<li> Use knowledge about light fields to smartly interpolate between existing rays!<\/li>\n<\/ul>\n<h4>Gaussian textures<\/h4>\n<ul>\n<li> Drop the phase (use random) of a texture, then model the spectrum with Gabor noise<\/li>\n<li>What about peripheral vision? Can one make a difference between Gaussian and non-gaussian textures?<\/li>\n<li>What about animation? The patterns obtained by varying phase but keeping spectrum might be interesting&#8230;<\/li>\n<\/ul>\n<h4>Diffusion textures<\/h4>\n<ul>\n<li> Use a boundary alternative to Laplace Equation (Green function)<\/li>\n<li>But requires culling to discard non-visible surfaces (really ad-hoc)<\/li>\n<\/ul>\n<h4>Micro-perceptual computation<\/h4>\n<ul>\n<li> Use humans to perform small micro-tasks without them knowing the overall plan<\/li>\n<\/ul>\n<h4>Wave BSDF<\/h4>\n<ul>\n<li> Transfer wave effects of light to geometric optics using a Wigner transform<\/li>\n<li>It outputs special rays that &#8220;cancel&#8221; light in places where interference is supposed to happen<\/li>\n<\/ul>\n<h4>Sky domes<\/h4>\n<ul>\n<li> Model sun and sky irradiance based on turbidity, sun location (w.r.t.time of day) and even ground albedo<\/li>\n<li>Use it to study shape perception? Clouds may convey more variations though&#8230;<\/li>\n<\/ul>\n<h4>Binocular tone mapping<\/h4>\n<ul>\n<li> Visual enrichment via binocular vision, applied to varying luminances.<\/li>\n<li>Use same approach to convey curvature via disparity?<\/li>\n<li>To the contrary, use rivalry to display different information (like transparency or IR, etc) on top of color image?<\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Schelling Points They claim that other descriptors are poorly related to theirs Same with GLS? Functional maps See Conformal flattening [Lipman] Use Laplace-Beltrami eigen values as functions Try with GLS? Sketch-based retrieval Use Gabor filters to characterize line-based renderings of 3D objects from multiple views Use quantization to &#8220;reduce&#8221; to a 1000 dim-descriptor Try replacing &#8230; <a title=\"Siggraph 2012\" class=\"read-more\" href=\"https:\/\/www.labri.fr\/perso\/barla\/blog\/?p=23531\" aria-label=\"Read more about Siggraph 2012\">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":[81],"tags":[],"class_list":["post-23531","post","type-post","status-publish","format-standard","hentry","category-talks"],"_links":{"self":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/23531","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=23531"}],"version-history":[{"count":1,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/23531\/revisions"}],"predecessor-version":[{"id":23541,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=\/wp\/v2\/posts\/23531\/revisions\/23541"}],"wp:attachment":[{"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=23531"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=23531"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.labri.fr\/perso\/barla\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=23531"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}