{"id":19484970,"url":"https://github.com/opencog/destin","last_synced_at":"2026-02-27T21:06:04.632Z","repository":{"id":10566673,"uuid":"12770895","full_name":"opencog/destin","owner":"opencog","description":"Deep SpatioTemporal Inference Network (DeSTIN) for vision processing","archived":false,"fork":false,"pushed_at":"2016-09-10T16:29:20.000Z","size":10510,"stargazers_count":34,"open_issues_count":1,"forks_count":28,"subscribers_count":24,"default_branch":"master","last_synced_at":"2023-10-20T22:12:02.825Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"http://wiki.opencog.org/w/DeSTIN","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"lgpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/opencog.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2013-09-12T00:05:02.000Z","updated_at":"2023-10-20T22:12:02.826Z","dependencies_parsed_at":"2022-09-03T17:11:38.407Z","dependency_job_id":null,"html_url":"https://github.com/opencog/destin","commit_stats":null,"previous_names":[],"tags_count":0,"template":null,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fdestin","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fdestin/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fdestin/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fdestin/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/opencog","download_url":"https://codeload.github.com/opencog/destin/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224009958,"owners_count":17240651,"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":"2024-11-10T20:24:56.383Z","updated_at":"2026-02-27T21:05:59.605Z","avatar_url":"https://github.com/opencog.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"DeSTIN\n=============\n\nPapers about DeSTIN\n-------------------\n* http://web.eecs.utk.edu/~itamar/Papers/BICA2009.pdf\n* http://www.ece.utk.edu/~itamar/Papers/BICA2011T.pdf\n* http://web.eecs.utk.edu/~itamar/Papers/AI_MAG_2011.pdf\n* http://research.microsoft.com/en-us/um/people/dongyu/nips2009/papers/Arel-DeSTIN_NIPS%20tpk2.pdf\n* http://goertzel.org/Goertzel_AAAI11.pdf\n* http://goertzel.org/DeSTIN_OpenCog_paper_v1.pdf\n* http://goertzel.org/papers/DeSTIN_OpenCog_paper_v2.pdf\n* http://www.springerlink.com/content/264p486742666751/fulltext.pdf\n* http://goertzel.org/VisualAttention_AGI_11.pdf\n* http://goertzel.org/Uniform_DeSTIN_paper.pdf\n* http://goertzel.org/papers/Uniform_DeSTIN_paper_v2.pdf\n* http://goertzel.org/papers/CogPrime_Overview_Paper.pdf\n\nCurrent Work\n------------\nImplementing uniform destin described here:\nhttp://wiki.opencog.org/w/DestinOpenCog\n\nCreating enhanced visualization capabilities.\n\nFor more detailed development progress see Diary.md\n\nBuilding\n--------\nInstructions below are for Ubuntu. \n\nSee Windows build instructions at https://github.com/opencog/destin/blob/master/WindowsBuild.md\n\nYou may build docker image containing all dependencies apart from CUDA SDK by using https://github.com/opencog/destin/blob/master/Dockerfile\n\nDependencies:\n\n* CMake (2.8.x, \u003e= 3.2)\n* OpenCV 2 (libopencv-dev)\n\nJava Bindings: JDK, ant and SWIG 2.x\n\nPython Bindings: python-dev (tested with python 2.7), python-opencv, and SWIG 2.x\n\nTo build DrentheDestin, CUDA SDK is required. \n\nIndividual parts can be skipped by commenting out their ADD_SUBDIRECTORY line in the CMakeLists.txt\nSome examples:\n\n* If you dont have CUDA for example, comment out the \"ADD_SUBDIRECTORY(DrentheDestin)\" in Destin/CMakeLists.txt\n\n* If you dont want to install any language bindings comment out \"ADD_SUBDIRECTORY(Bindings)\" in Destin/CMakeLists.txt.\n\n* If you want python bindings but not Java, then in Destin/Bindings/CMakeLists.txt comment out \"add_subdirectory(Java)\"\n\nBuilding:\n\n    $ git clone http://github.com/opencog/destin.git\n    $ cd destin\n    $ git submodule init\n    $ git submodule update\n    $ cd Destin\n    $ cmake . \n    $ make\n    \nRebuilding:\n    \nDuring development sometimes CMake is not able to detect dependencies between the different built libraries. So it is recommened\nto do a \"make clean\" and \"make -j4\" (j4 lets you use multiple cores) from the Destin directory to make sure everyting is building fresh.\n\nBuilding Java Bindings:\n\nThe native jni bindings lib is built with CMake, the actual java program which uses it is built with java_build.sh:\n\n    $ cd Destin/Bindings/Java\n    $ ./java_build.sh\n    $ ./java_run.sh\n    \n\nShow me something!\n------------------\n\nCurrently the most interesting things to see are the scripts in the Destin/Bindings/Python directory. \nThese scripts are meant to be used interactivly so I recommend installing and using idle for python ( search for and install the idle package).\n\nTo see a self organizing map (SOM):\n\nManually download http://www.cs.toronto.edu/~kriz/cifar-10-binary.tar.gz\n\n    $ tar xf cifar-10-binary.tar.gz\n    $ cd cifar-10-batches-bin\n    $ pwd\n\nNote the absolute path \n\n    $ cd Destin/Bindings/Python\n    \nEdit som.py and edit the cifar_dir variable near the top and change it to the absolute path of the cifar-10-batches-bin directory\n\n    $ idle som.py\n\nTwo windows appear, select the som.py window and click Run-\u003eRun Module (F5).\nTraining will begin, after less than 5 minutes the SOM will popup.\n    \nTo see DeSTIN train on an video and see its output:\n    \n    $ cd Destin/Bindings/Python\n    $ idle dostuff.py\n    \nTwo windows should appear.\n\n    - select the dostuff.py window\n    - click Run -\u003e Run Modedule ( F5 )\n\nYou should see a video of a hand and windows for 8 layers.\n    \nIn the other \"*Python Shell*\" window you can interact with the code after its done with 500 frames ( about a minute).\nTo make it continue processing type go(100) to have it process 100 more frames. You can make it train on webcam input instead by\nchanging the line: \n\n    vs = pd.VideoSource(False, \"hand.m4v\")\n    \nto \n\n    vs = pd.VideoSource(True, \"\")\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopencog%2Fdestin","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fopencog%2Fdestin","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopencog%2Fdestin/lists"}