{"url":"/dataset/vizdoom","name":"VizDoom","full_name":"VizDoom","description_markdown":"ViZDoom is an AI research platform based on the classical First Person Shooter game Doom. The most popular game mode is probably the so-called Death Match, where several players join in a maze and fight against each other. After a fixed time, the match ends and all the players are ranked by the FRAG scores defined as kills minus suicides. During the game, each player can access various observations, including the first-person view screen pixels, the corresponding depth-map and segmentation-map (pixel-wise object labels), the bird-view maze map, etc. The valid actions include almost all the keyboard-stroke and mouse-control a human player can take, accounting for moving, turning, jumping, shooting, changing weapon, etc. ViZDoom can run a game either synchronously or asynchronously, indicating whether the game core waits until all players’ actions are collected or runs in a constant frame rate without waiting.\r\n\r\nSource: [Arena: a toolkit for Multi-Agent Reinforcement Learning](https://arxiv.org/abs/1907.09467)\r\nImage Source: [https://github.com/mwydmuch/ViZDoom](https://github.com/mwydmuch/ViZDoom)","description_withheld":null,"homepage":"http://vizdoom.cs.put.edu.pl/","introduced_date":"2016-01-01","introduced_date_note":null,"introduced_by":{"paper":"/paper/vizdoom-a-doom-based-ai-research-platform-for","title":"ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning","first_author":"Michał Kempka","url":null},"license":{"name":"Custom (multiple)","url":"https://github.com/mwydmuch/ViZDoom#license"},"modalities":[{"name":"Environment","url":"/datasets/modality/environment"}],"tasks":[{"name":"Image Generation","url":"/task/image-generation","datasets_with_task":"/datasets/task/image-generation"},{"name":"Scene Generation","url":"/task/scene-generation","datasets_with_task":"/datasets/task/scene-generation"},{"name":"Decision Making","url":"/task/decision-making","datasets_with_task":"/datasets/task/decision-making"},{"name":"Game of Doom","url":"/task/game-of-doom","datasets_with_task":"/datasets/task/game-of-doom"},{"name":"Q-Learning","url":"/task/q-learning","datasets_with_task":"/datasets/task/q-learning"},{"name":"Starcraft","url":"/task/starcraft","datasets_with_task":"/datasets/task/starcraft"}],"languages":[],"variants":["ViZDoom Basic Scenario","VizDoom"],"data_loaders":[],"num_papers_in_archive":156,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/image-generation-on-vizdoom","task":"Image Generation","dataset_variant":"VizDoom","rows":4,"metrics":["FID","FID (SwAV)"],"first_row_in_archive_order":{"model":"GAUDI","paper":"/paper/gaudi-a-neural-architect-for-immersive-3d","metrics":{"FID":"33.7","FID (SwAV)":"3.24"},"code_links":[{"title":"apple/ml-gaudi","url":"https://github.com/apple/ml-gaudi"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/scene-generation-on-vizdoom","task":"Scene Generation","dataset_variant":"VizDoom","rows":3,"metrics":["FID","SwAV-FID"],"first_row_in_archive_order":{"model":"GSN","paper":"/paper/unconstrained-scene-generation-with-locally","metrics":{"FID":"37.21","SwAV-FID":"4.56"},"code_links":[{"title":"apple/ml-gsn","url":"https://github.com/apple/ml-gsn"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/game-of-doom-on-vizdoom-basic-scenario","task":"Game of Doom","dataset_variant":"ViZDoom Basic Scenario","rows":1,"metrics":["Average Score"],"first_row_in_archive_order":{"model":"DQN","paper":"/paper/vizdoom-a-doom-based-ai-research-platform-for","metrics":{"Average Score":"82.2"},"code_links":[{"title":"NervanaSystems/coach","url":"https://github.com/NervanaSystems/coach"},{"title":"mwydmuch/ViZDoom","url":"https://github.com/mwydmuch/ViZDoom"},{"title":"farama-foundation/vizdoom","url":"https://github.com/farama-foundation/vizdoom"},{"title":"apollopower/DOOM-AI","url":"https://github.com/apollopower/DOOM-AI"},{"title":"hegde95/ViZDoom_with_Sound","url":"https://github.com/hegde95/ViZDoom_with_Sound"},{"title":"nolanwinsman/Team-Doom","url":"https://github.com/nolanwinsman/Team-Doom"},{"title":"chengyu2/vizdoom_rl_community_canberra","url":"https://github.com/chengyu2/vizdoom_rl_community_canberra"},{"title":"icmlanon58443043/vizdoomicmlanon","url":"https://github.com/icmlanon58443043/vizdoomicmlanon"},{"title":"icmlanon58443043/vizdoom","url":"https://github.com/icmlanon58443043/vizdoom"},{"title":"sagpant/ViZDoom","url":"https://github.com/sagpant/ViZDoom"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/gaudi-a-neural-architect-for-immersive-3d","title":"GAUDI: A Neural Architect for Immersive 3D Scene Generation","date":"2022-07-27","rows_on_this_dataset":4,"code_links":1,"syntology":null},{"paper":"/paper/unconstrained-scene-generation-with-locally","title":"Unconstrained Scene Generation with Locally Conditioned Radiance Fields","date":"2021-04-01","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/pi-gan-periodic-implicit-generative","title":"pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis","date":"2020-12-02","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":3,"samples_unverified":1,"pointer_only_for_licence":2,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/graf-generative-radiance-fields-for-3d-aware","title":"GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis","date":"2020-07-05","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":12,"samples_ran":3,"samples_unverified":9,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/vizdoom-a-doom-based-ai-research-platform-for","title":"ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning","date":"2016-05-06","rows_on_this_dataset":1,"code_links":10,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":2,"samples_unverified":3,"pointer_only_for_licence":5,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":3,"samples_harvested":21,"samples_ran":8,"samples_unverified":13,"pointer_only_for_licence":7,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}