{"id":19112732,"url":"https://github.com/pgrit/var-aware-mis-pbrt","last_synced_at":"2025-04-30T22:14:47.163Z","repository":{"id":49834369,"uuid":"167016130","full_name":"pgrit/var-aware-mis-pbrt","owner":"pgrit","description":"Implementation of the paper \"Variance-Aware Multiple Importance Sampling\" for bidirectional path tracing in PBRT.","archived":false,"fork":false,"pushed_at":"2019-08-28T10:28:53.000Z","size":5818,"stargazers_count":36,"open_issues_count":0,"forks_count":5,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-04-30T22:14:41.531Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-2-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/pgrit.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-01-22T15:19:12.000Z","updated_at":"2024-10-17T03:54:41.000Z","dependencies_parsed_at":"2022-09-05T10:31:16.064Z","dependency_job_id":null,"html_url":"https://github.com/pgrit/var-aware-mis-pbrt","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pgrit%2Fvar-aware-mis-pbrt","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pgrit%2Fvar-aware-mis-pbrt/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pgrit%2Fvar-aware-mis-pbrt/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pgrit%2Fvar-aware-mis-pbrt/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pgrit","download_url":"https://codeload.github.com/pgrit/var-aware-mis-pbrt/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251789618,"owners_count":21644086,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-09T04:33:59.027Z","updated_at":"2025-04-30T22:14:47.140Z","avatar_url":"https://github.com/pgrit.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"Variance-aware MIS weights in PBRTv3\n===================================================\n\nThis repository implements the approach discussed in the\n\"Variance-Aware Multiple Importance Sampling\" paper.\n\nThe core of the implementation is in `src/util/samis.h` and `src/util/samis.cpp`:\nThe class `SAMISRectifier` implements the rectification as a black box: given the outcome of paths\nsampled from the first iteration, it computes the required variance estimates and the resulting factors.\n\nFor reference value computations, the files `src/util/varestim.h` and `src/util/varestim.cpp` provide a utility\nclass to compute accurate estimates of the variance factors given a large number of samples.\n\nThe implementation required minor changes in the bidirectional path tracer integrator\n(`src/integrators/bdpt.cpp` and `bdpt.h`) to separate the rendering into multiple iterations\nand to look-up and multiply with the propper variance factors during the MIS computation.\nFurthermore, the weighted combination of the first iteration with the following one also\nrequired a small addition to the `Film` class in `src/core/film.cpp` and `src/core/film.h`.\nThe defensive sampling application is implemented in a new integrator, see `src/integrators/guideddi.cpp` and\n`src/integrators/guideddi.h`. The implementation is analogous to the one from the \"Optimal Multiple Importance Sampling\"\npaper by Kondapaneni et al.\n\nFor build instructions, documentation, test scenes, etc., refer to [the original PBRTv3 repository](https://github.com/mmp/pbrt-v3/)\nor the repository created by [Benedikt Bitterli](https://benedikt-bitterli.me/resources/).\nAll results in the paper were generated from (sometimes slightly modified versions of) the scenes from those repositories.\n\nTesting\n---------\n\nThe folder `test` contains a number of Python scripts to render comparison images with various approaches.\n\nThe `test/runtests_bdpt.py` script loads a number of PBRT scene files (specified in the script) and replaces\nthe integrator and sampler definitions by the appropriate methods, as given in the beginning of the script.\n\nThe `test/runtests_di.py` script does the same for the defensive sampling application. For the comparison to the\noptimal MIS weights, the path to the source code from that paper needs to be hard-coded in the script (by modifying\nthe `optimal_mis_executable` variable in the beginning.)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpgrit%2Fvar-aware-mis-pbrt","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpgrit%2Fvar-aware-mis-pbrt","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpgrit%2Fvar-aware-mis-pbrt/lists"}