{"id":19199844,"url":"https://github.com/rose-stl-lab/ir2rgb","last_synced_at":"2025-04-10T11:23:34.533Z","repository":{"id":114950179,"uuid":"224539916","full_name":"Rose-STL-Lab/ir2rgb","owner":"Rose-STL-Lab","description":"IR to RGB video translation","archived":false,"fork":false,"pushed_at":"2023-07-11T04:14:03.000Z","size":14559,"stargazers_count":3,"open_issues_count":1,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-03-24T10:11:14.543Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Rose-STL-Lab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-11-28T00:29:36.000Z","updated_at":"2024-03-14T11:20:59.000Z","dependencies_parsed_at":"2024-11-09T12:29:53.828Z","dependency_job_id":"cf0a56fb-35d0-4437-9349-79f79f1c7b66","html_url":"https://github.com/Rose-STL-Lab/ir2rgb","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/Rose-STL-Lab%2Fir2rgb","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Rose-STL-Lab%2Fir2rgb/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Rose-STL-Lab%2Fir2rgb/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Rose-STL-Lab%2Fir2rgb/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Rose-STL-Lab","download_url":"https://codeload.github.com/Rose-STL-Lab/ir2rgb/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248208608,"owners_count":21065203,"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":"2024-11-09T12:29:10.949Z","updated_at":"2025-04-10T11:23:34.508Z","avatar_url":"https://github.com/Rose-STL-Lab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Paper from Traffic4Cast and IR 2 RGB video translation\nWrite-up and findings here: https://github.com/Rose-STL-Lab/ir2rgb/blob/master/DL%20Methods%20for%20Video%20Prediction.pdf\n\n# IR 2 RGB video translation\n\nFrom vid2vid (https://github.com/NVIDIA/vid2vid)\n\n## Prerequisites\n- Linux or macOS\n- Python 3\n- GCC version \u003c= 6\n- NVIDIA GPU + CUDA cuDNN\n- PyTorch 0.4+\n\n## Additional downloads \n- The file `FlowNet2_checkpoint.pth.tar` is too large to upload to github and needs to be downloaded seperately. It contains the FlowNet2 model and is used to construct the flow model for vid2vid\n- To download the model run `python scripts/download_models_flownet2`. Now try training the ir2rgb network.\n- If you have any additional errors and need to compile a snapshot of the FlowNet2:\n\t- First try with the scripts I made for this repo. Go to the location `ir2rgb/models/flownet2_pytorch/` and run the command line: `bash install.sh` (or run each command inside `install.sh` in the command line individually).\n\t\t- If your gpu's are not accounted for in this compilation, you might have to add their arch codes: Add the line '-gencode', 'arch=compute_XX,code=sm_XX', (where XX is GPU specific) in all 3 setup.py scripts in `channelnorm_package`, `correlation_package`, and `resample2d_package`. Then install flownet.\n\t- If even this doesn't work, you might have install it from scratch: first go to: https://docs.google.com/uc?export=download\u0026id=1gKwE1Ad41TwtAzwDcN3dYa_S6DcVyiSl and download `flownet2_pytorch_041.zip`. To compile:\n\t\t- Unzip and copy the entire content of the extracted `flownet2_pytorch_041.zip` file to the location `ir2rgb/models/` (and replace any existing files with the same names)\n\t\t- Then go to the location `ir2rgb/models/flownet2_pytorch/` and run the command line: `bash install.sh` (or run each command in `install.sh` in the command line individually)\n\t- For more information please see: https://github.com/NVIDIA/vid2vid/issues/86 and https://github.com/NVIDIA/vid2vid/issues/82 \n\n### Training \n\n- To train the model, from the command line run\n    ```\n    python train.py --name Name_of_training_session --dataroot sample_dataset --input_nc 3 --output_nc 3 --tf_log\n    ```\n  - Or, simply bash the script (which contains the command line)\n    ```\n    bash run_to_train.sh\n    ```  \n### Testing\n\n- To test the model, from the command line run\n    ```\n    python test.py --name name_of_test_session --dataroot sample_dataset --input_nc 3 --output_nc 3 --use_real_img\n    ```\n  - Or, simply bash the script (which contains the command line)\n    ```\n    bash run_to_test.sh\n    ```  \n- For a detailed list of options, please see the options folder\n\n### Dataset\n- Dataset folder must be in the following format: \n\n\t```\n\t.\n\t├── dataset_folder    \n\t\t│\n\t\t├── train_A    # input training IR images\n\t\t├── train_B    # output training RGB images\n\t\t├── test_A     # input test IR images\n\t\t└── test_B     # output test RGB images\n\t```\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frose-stl-lab%2Fir2rgb","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Frose-stl-lab%2Fir2rgb","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frose-stl-lab%2Fir2rgb/lists"}