{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":592828987,"defaultBranch":"main","name":"zimt","ownerLogin":"kfjahnke","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-01-24T16:16:56.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/28724791?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1674577017.428211","currentOid":""},"activityList":{"items":[{"before":"cae9b77c308d3dd8e6d85297756b3e475e222295","after":"581589bd17e3dd4f44ce51d130f181f79c2068c8","ref":"refs/heads/main","pushedAt":"2024-05-26T09:58:19.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added examples for the three filter types: iir.cc, fir.cc and prefilter.cc","shortMessageHtmlLink":"added examples for the three filter types: iir.cc, fir.cc and prefilt…"}},{"before":"b6fafdb33834be90311e7d8a96a7677817774394","after":"cae9b77c308d3dd8e6d85297756b3e475e222295","ref":"refs/heads/main","pushedAt":"2024-05-26T09:43:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"this commit adds code for digital filters:\nI've ported vspline's digital filtering code to work with zimt data types instead of\nusing vigra. The filters are quite specific: separable convolution, and forward-backward\nn-pole recursive filters - the latter with a specialization for b-spline prefiltering.\nThe filtering code is not used by other parts of zimt - to use the filters, #include\nconvolve.h, recursive.h or prefilter.h and use the function templates given at the\nend of these files. While the recursive filters are quite specific to b-spline\nprocessing, they can be useful as effective low-pass-filters (use one positive\npole under 1.0). Separable convolution is probably more generally useful.","shortMessageHtmlLink":"this commit adds code for digital filters:"}},{"before":"e3ed3d9bcfed492e742c0e991c2ee65444df5a7b","after":"b6fafdb33834be90311e7d8a96a7677817774394","ref":"refs/heads/main","pushedAt":"2024-04-19T07:22:09.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added code to 'twine' the interpolation:\nthis is a new idea: the 'twining' wrapper around the interpolator results in\nseveral pick-ups in the vicinity of the pick-up point and subsequent aggregation,\nwhich - with the parameterization used here - has the same effect as  oversampling\nand subsequent box-filtering. The concept is orthogonal to the interpolator used,\nbut when OIIO's antialiasing filter is used for pick-up, it's quite pointless\nto add 'twining', so it's best done with itp == 1.","shortMessageHtmlLink":"added code to 'twine' the interpolation:"}},{"before":"025e1c487c1d3fcd3a21961e73d3612bbf32a168","after":"e3ed3d9bcfed492e742c0e991c2ee65444df5a7b","ref":"refs/heads/main","pushedAt":"2024-04-17T10:01:08.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added copyright notice to metrics.h","shortMessageHtmlLink":"added copyright notice to metrics.h"}},{"before":"202018e280ed918964f92f2264f82f5631fd8d12","after":"025e1c487c1d3fcd3a21961e73d3612bbf32a168","ref":"refs/heads/main","pushedAt":"2024-04-17T09:59:28.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"factored out metrics.c (from envutil.cc), added 'ctc' parameter to envutil:\nthe 'ctc' parameter indicates that the cube face fov is measured between\nmarginal pixel centers, not pixel edges, which is the default.","shortMessageHtmlLink":"factored out metrics.c (from envutil.cc), added 'ctc' parameter to en…"}},{"before":"2d368a2f983136a252b1587bde0ea2a1e68566a3","after":"202018e280ed918964f92f2264f82f5631fd8d12","ref":"refs/heads/main","pushedAt":"2024-04-16T08:07:14.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"scanned and modified some of the initial comments","shortMessageHtmlLink":"scanned and modified some of the initial comments"}},{"before":"d9dd252d8d75852570cef6fcaf35ea1d422f7179","after":"2d368a2f983136a252b1587bde0ea2a1e68566a3","ref":"refs/heads/main","pushedAt":"2024-04-12T15:40:02.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"grooming latlon.cc, minor bug fixes, handling of tile-based input:\nwhen feeding an openEXR cubemap, the program crashed on me - the cubemap\nheld tile-based data, and the scanline-based read failed, strangely\nwith a floating point exception. Now the program looks whether the\ninput supports scanlines and if not, reads the entire image into a\nbuffer, then copies the six cube faces to the IR image.","shortMessageHtmlLink":"grooming latlon.cc, minor bug fixes, handling of tile-based input:"}},{"before":"c0684b4a528fc71e566cd7efde0d2b022e6e1edc","after":"d9dd252d8d75852570cef6fcaf35ea1d422f7179","ref":"refs/heads/main","pushedAt":"2024-04-11T17:17:21.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"fixed bug which prevented use of anisotropic filter for texture lookup","shortMessageHtmlLink":"fixed bug which prevented use of anisotropic filter for texture lookup"}},{"before":"5a0562860df949b90acb0d7cf00bd2e930e8bf1c","after":"c0684b4a528fc71e566cd7efde0d2b022e6e1edc","ref":"refs/heads/main","pushedAt":"2024-04-10T09:54:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"grooming latlon.cc","shortMessageHtmlLink":"grooming latlon.cc"}},{"before":"19e8ecc8a7250bb87bf7213d3bd5b197da324dfc","after":"5a0562860df949b90acb0d7cf00bd2e930e8bf1c","ref":"refs/heads/main","pushedAt":"2024-04-09T19:23:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"using TextureOpt::MipModeTrilinear to avoid black-lane bug:\nmore trials have hinted that the default mipmapping mode (MipModeAniso)\nmay be the only one producing the bug which has kept me from employing OIIO\nfor texture lookup in latlon.cc - that's annoying, because it's both\nthe default and the one promising the best result. so for now I'm switching\nto MipModeTrilinear, which is probably the next best one.","shortMessageHtmlLink":"using TextureOpt::MipModeTrilinear to avoid black-lane bug:"}},{"before":"6c0b53f64b7771dae323e9135acaedbcb6dc4fd7","after":"19e8ecc8a7250bb87bf7213d3bd5b197da324dfc","ref":"refs/heads/main","pushedAt":"2024-04-08T08:05:23.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"bug fix in several SIMD back-ends, disusing OIIO code for texture lookup:\nthe bug fix removes problems with g++ and using binary operators with\nfundamentals on the lhs and SIMD types on the rhs. These are coded as\nfriend functions, and g++ did not accept the using definition for vsize\nI had in place, whereas clang++ did. The fixed version works with both\ncompilers.\nI am having problems with OIIO's texture lookup, which I'd like to use\nto glean pixel data from the IR image (the image with the extra frame\nof support pixels around the cube face images). With OIIO's 'texture'\nfunction, I should be able to get the pixel data with proper mip-mapping\nand an anisotropic antialiasing filter, but currently I can't get it\nto work properly. The code is all there, but instead I'm routing to\nsimple bilinear interpolation directly from the IR image. This produces\nokay results if there isn't too much of a scale change involved.","shortMessageHtmlLink":"bug fix in several SIMD back-ends, disusing OIIO code for texture loo…"}},{"before":"7f3d22b3c1579329c9456170c6ea57a88765b4f9","after":"6c0b53f64b7771dae323e9135acaedbcb6dc4fd7","ref":"refs/heads/main","pushedAt":"2024-04-05T09:59:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added texture lookup to latlon.cc:\nthis is a bit awkward for now and needs a temporary image.","shortMessageHtmlLink":"added texture lookup to latlon.cc:"}},{"before":"a2dd548d0d9a44acd3ec4dd9f2fe95da3af331a1","after":"7f3d22b3c1579329c9456170c6ea57a88765b4f9","ref":"refs/heads/main","pushedAt":"2024-03-31T20:04:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added traits class zimt::is_integral, tweaked latlon.cc:\nThe traits class allows the operator code to recognize simdized types\nand their aggregates as consisting of integral components. In latlon.cc\nI moved to using 3D ray coordinates in pixel units, omitting the\nscaling-down to model space units. This makes it possible to keep\nthe calculation in integer just up to the unavoidable division to\nproject the rays onto a plane, which may be faster and produces\nmore procise results.","shortMessageHtmlLink":"added traits class zimt::is_integral, tweaked latlon.cc:"}},{"before":"4179c5dc042a11b37017511867a7c4c6af1b639e","after":"a2dd548d0d9a44acd3ec4dd9f2fe95da3af331a1","ref":"refs/heads/main","pushedAt":"2024-03-28T10:50:14.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"some refactoring, support of RGBA, monochrome:\nSome of the functors moved into the sixfold_t object itself.\nThe channel count is now a template, and the program handles RGBA\nand monochrome files on top of plain RGB.","shortMessageHtmlLink":"some refactoring, support of RGBA, monochrome:"}},{"before":"1ca943101c9d18ddb34c870052146af7aefb7e77","after":"4179c5dc042a11b37017511867a7c4c6af1b639e","ref":"refs/heads/main","pushedAt":"2024-03-26T08:45:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"polished latlon.cc, fixed bug which made some compilations with g++ fail","shortMessageHtmlLink":"polished latlon.cc, fixed bug which made some compilations with g++ fail"}},{"before":"dfe3dba364a64a6fc1f0fe9a5e05a479c5ec9297","after":"1ca943101c9d18ddb34c870052146af7aefb7e77","ref":"refs/heads/main","pushedAt":"2024-03-24T17:38:02.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"debugged and polished latlon.cc, using bilinear interpolation in cubemap.cc:\nlatlon.cc had the pickup coordinate off by 0.5. I added more comments,\nespecially the initial comment, which is much more detailed now. I also\ndid a bit of refactoring and tidying up. cubemap.cc was changed to use\nbilinear interpolation so I can compare it's output and performance with\na version I have which is not using OIIO (currently unpublished).","shortMessageHtmlLink":"debugged and polished latlon.cc, using bilinear interpolation in cube…"}},{"before":"03f79959d1aa9864ddcbb85f26997108ebcf6e79","after":"dfe3dba364a64a6fc1f0fe9a5e05a479c5ec9297","ref":"refs/heads/main","pushedAt":"2024-03-23T17:21:17.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"tweaked example program latlon.cc, added more comments","shortMessageHtmlLink":"tweaked example program latlon.cc, added more comments"}},{"before":"8fc191a2328e5bfde35774d828b8ac58b6fed6ab","after":"03f79959d1aa9864ddcbb85f26997108ebcf6e79","ref":"refs/heads/main","pushedAt":"2024-03-22T20:39:22.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"bug fix: storing result image failed due to wrong parameterization.","shortMessageHtmlLink":"bug fix: storing result image failed due to wrong parameterization."}},{"before":"cf3130b1cbfb6a4a9ecbbb5451e630181977f44c","after":"8fc191a2328e5bfde35774d828b8ac58b6fed6ab","ref":"refs/heads/main","pushedAt":"2024-03-22T20:24:37.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"small changes to zimt library code, elaborated example latlon.cc (ex skybox.cc)\nlatlon.cc now generates mathematically correct support around the cube\nfaces embedded in the internal representation and uses bilinear interpolation\nto produce the output.","shortMessageHtmlLink":"small changes to zimt library code, elaborated example latlon.cc (ex …"}},{"before":"3107e9d6b0f805ebe7bb5697ff3fc836bdedb27a","after":"cf3130b1cbfb6a4a9ecbbb5451e630181977f44c","ref":"refs/heads/main","pushedAt":"2024-03-21T09:39:32.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added example program 'skybox.cc' which calculates a lat/lon environment map from a cubemap:\nthis is a first working implementation, but it only obtains the pixel data with a\nsimple nearest-neighbour lookup and does not populate the 'support' pixels in\nthe internal representation to support better interpolators. Nevertheless this\ninitial commit demonstates that the geometry and logic work as intended.","shortMessageHtmlLink":"added example program 'skybox.cc' which calculates a lat/lon environm…"}},{"before":"29b2de57378678a87076a156024780a7699c3150","after":"3107e9d6b0f805ebe7bb5697ff3fc836bdedb27a","ref":"refs/heads/main","pushedAt":"2024-03-16T10:55:16.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added example program 'cubemap.cc' which creates an openEXR-compatible\ncubemap from a full spherical. This is based on the code in 'rectilinear.cc'\nand uses the same method to generate the six cube faces.","shortMessageHtmlLink":"added example program 'cubemap.cc' which creates an openEXR-compatible"}},{"before":"cc1566a36acee099fff71f78cd21e1db558b15f6","after":"29b2de57378678a87076a156024780a7699c3150","ref":"refs/heads/main","pushedAt":"2024-03-13T18:14:36.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"exported reductions of highway-based masks to namespace zimt, changed a bit of docu","shortMessageHtmlLink":"exported reductions of highway-based masks to namespace zimt, changed…"}},{"before":"19a492ba9d08c2a95c1cb6fec1fa6313ee9a7a8a","after":"cc1566a36acee099fff71f78cd21e1db558b15f6","ref":"refs/heads/main","pushedAt":"2024-03-12T10:36:18.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"small modifications for readability, comments in example rectilinear.cc","shortMessageHtmlLink":"small modifications for readability, comments in example rectilinear.cc"}},{"before":"5d6962bdfdddaf3033a8bea22763bb88a55608aa","after":"19a492ba9d08c2a95c1cb6fec1fa6313ee9a7a8a","ref":"refs/heads/main","pushedAt":"2024-03-12T09:02:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"brushed up rectilinear.cc, extensive comments, can use Vc and std::simd","shortMessageHtmlLink":"brushed up rectilinear.cc, extensive comments, can use Vc and std::simd"}},{"before":"88cc423c3e374dd952e5ff8551f214a321aa3e10","after":"5d6962bdfdddaf3033a8bea22763bb88a55608aa","ref":"refs/heads/main","pushedAt":"2024-03-06T18:13:46.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"brushed up rectilinear.cc, wrote more comments. It's turning useful.","shortMessageHtmlLink":"brushed up rectilinear.cc, wrote more comments. It's turning useful."}},{"before":"1f0279f80767a491d55e7664fd0527da5fe34bf2","after":"88cc423c3e374dd952e5ff8551f214a321aa3e10","ref":"refs/heads/main","pushedAt":"2024-03-05T11:00:06.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"changed example 'rectilinear.cc' to work with 3D grid coordinates\nand to take yaw, pich and roll on the command line. The program works\nand extracts correct linear images from a full spherical, but the code\nneeds some tidying up and comments.","shortMessageHtmlLink":"changed example 'rectilinear.cc' to work with 3D grid coordinates"}},{"before":"188b407ed5260b8b68e1db804c99a3d0f9e2bd7c","after":"1f0279f80767a491d55e7664fd0527da5fe34bf2","ref":"refs/heads/main","pushedAt":"2024-03-04T18:17:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"slight reformulation of getting the pointers to pass to 'environment'","shortMessageHtmlLink":"slight reformulation of getting the pointers to pass to 'environment'"}},{"before":"61793501ef328cc701e91cd7f34b9569987f1c91","after":"188b407ed5260b8b68e1db804c99a3d0f9e2bd7c","ref":"refs/heads/main","pushedAt":"2024-03-04T18:15:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"proper calculation of derivatives for environment lookup in rectilinear.cc","shortMessageHtmlLink":"proper calculation of derivatives for environment lookup in rectiline…"}},{"before":"c7ab87beb9a60c4480a1e35f611e258ae3d42403","after":"61793501ef328cc701e91cd7f34b9569987f1c91","ref":"refs/heads/main","pushedAt":"2024-03-03T11:29:13.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"elaborated example program rectilinear.cc, proposed some options for environment lookup","shortMessageHtmlLink":"elaborated example program rectilinear.cc, proposed some options for …"}},{"before":"002d0eecdfe863299398cdcb321bed6671631105","after":"c7ab87beb9a60c4480a1e35f611e258ae3d42403","ref":"refs/heads/main","pushedAt":"2024-03-02T11:43:18.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"kfjahnke","name":null,"path":"/kfjahnke","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/28724791?s=80&v=4"},"commit":{"message":"added first draft of example 'rectilinear.cc', and some minor changes like typo fixes.","shortMessageHtmlLink":"added first draft of example 'rectilinear.cc', and some minor changes…"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEVBVYCAA","startCursor":null,"endCursor":null}},"title":"Activity · kfjahnke/zimt"}