{"id":21390751,"url":"https://github.com/interpause/dinosavi","last_synced_at":"2025-03-16T13:17:16.385Z","repository":{"id":166383925,"uuid":"601869124","full_name":"Interpause/dinosavi","owner":"Interpause","description":"Self-supervised learning of Video Object Segmentation using DINOSAUR and SAVi","archived":false,"fork":false,"pushed_at":"2025-01-05T11:23:16.000Z","size":1110,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-23T01:08:51.427Z","etag":null,"topics":["computer-vision","object-centric-learning","object-centric-video-prediction","object-segmentation","video-object-segmentation"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/Interpause.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":"2023-02-15T01:57:17.000Z","updated_at":"2025-01-05T11:23:19.000Z","dependencies_parsed_at":null,"dependency_job_id":"690e1317-c711-4b0d-a9e4-17d5df5918b0","html_url":"https://github.com/Interpause/dinosavi","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/Interpause%2Fdinosavi","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Interpause%2Fdinosavi/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Interpause%2Fdinosavi/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Interpause%2Fdinosavi/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Interpause","download_url":"https://codeload.github.com/Interpause/dinosavi/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243871914,"owners_count":20361380,"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":["computer-vision","object-centric-learning","object-centric-video-prediction","object-segmentation","video-object-segmentation"],"created_at":"2024-11-22T13:19:36.046Z","updated_at":"2025-03-16T13:17:16.358Z","avatar_url":"https://github.com/Interpause.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DINOSAVi\n\n\u003e Self-supervised learning of Video Object Segmentation using DINOSAUR and SAVi\n\n**Presentation Slides**: \u003chttps://1drv.ms/p/s!AgE9E4ZerfvahsdfaYUe8caynzF0iw\u003e\n\n![slot-s256d-line-t01-tline-ini5](https://github.com/Interpause/dinosavi/assets/42513874/625d61a7-6f5f-4f43-aa4d-f11d812eef43)\n\u003e VOS shown above is from `slot-s256d-line-t01-tline-ini5` model\n\nModels: \u003chttps://1drv.ms/f/s!AgE9E4Zerfvahsd8TTjyDmHvikwY6g\u003e\n\n**Note**: \u003chttps://hydra.cc/\u003e is used to manage configuration, CLI and experiment running. See \u003chttps://hydra.cc/docs/advanced/override_grammar/basic/\u003e for CLI override grammar. Jump to [Codebase Notes](#codebase-notes) for more info.\n\n## Installation\n\n```sh\n# (Optional) Use conda for virtual environment instead. Poetry creates venv by default.\nconda create -n dinosavi python=3.10\npoetry install\n```\n\n## Training\n\n```sh\npython -m dinosavi mode=train hparam={hparam_file_name} exp_name={experiment_name}\n```\n\n## Evaluation\n\n```sh\npython -m dinosavi mode=eval resume={.ckpt_to_load} exp_name={results_name} device={cpu_or_cuda}\n```\n\n## Codebase Notes\n\n- [Hydra](https://hydra.cc/) (built on [OmegaConf](https://omegaconf.readthedocs.io/)) is used for configuration management, CLI and experiment running.\n  - For CLI options, you can refer to [`dinosavi/cfg/main.yaml`](dinosavi/cfg/main.yaml) and [`dinosavi/cfg/mode`](dinosavi/cfg/mode/).\n  - Hyperparameter config files are stored in [`dinosavi/cfg/hparam`](dinosavi/cfg/hparam/).\n    - [OmegaConf Variable Interpolation](https://omegaconf.readthedocs.io/en/2.3_branch/usage.html#variable-interpolation) is heavily used to link hyperparameter values into relevant config files.\n  - Many parts of the code (i.e., models, datasets, trainers) are implemented as config files using [Hydra's Instantiate API](https://hydra.cc/docs/advanced/instantiate_objects/overview/).\n- We use [Black](https://github.com/psf/black) for formatting, [isort](https://github.com/PyCQA/isort) for import sorting, and Google-style docstrings.\n- Anything to do with [Contrastive Random Walk](https://ajabri.github.io/videowalk/) (CRW) is leftover legacy code from earlier experiments.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Finterpause%2Fdinosavi","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Finterpause%2Fdinosavi","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Finterpause%2Fdinosavi/lists"}