{"url":"/dataset/vidstg","name":"VidSTG","full_name":null,"description_markdown":"The **VidSTG** dataset is a spatio-temporal video grounding dataset constructed based on the video relation dataset VidOR. VidOR contains 7,000, 835 and 2,165 videos for training, validation and testing, respectively. The goal of the Spatio-Temporal Video Grounding task (STVG) is to localize the spatio-temporal section of an untrimmed video that matches a given sentence depicting an object. **VidSTG** contains 5,563, 618, and 743 videos for training, validation, and testing, respectively. \r\n\r\nSource: [https://github.com/Guaranteer/VidSTG-Dataset](https://github.com/Guaranteer/VidSTG-Dataset)\r\nImage Source: [https://github.com/Guaranteer/VidSTG-Dataset](https://github.com/Guaranteer/VidSTG-Dataset)","description_withheld":null,"homepage":"https://github.com/Guaranteer/VidSTG-Dataset","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/where-does-it-exist-spatio-temporal-video","title":"Where Does It Exist: Spatio-Temporal Video Grounding for Multi-Form Sentences","first_author":"Zhu Zhang","url":null},"license":null,"modalities":[],"tasks":[{"name":"Spatio-Temporal Video Grounding","url":"/task/spatio-temporal-video-grounding","datasets_with_task":"/datasets/task/spatio-temporal-video-grounding"},{"name":"Temporal Localization","url":"/task/temporal-localization","datasets_with_task":"/datasets/task/temporal-localization"}],"languages":[],"variants":["VidSTG"],"data_loaders":[{"repo":"https://github.com/Guaranteer/VidSTG-Dataset","url":"https://github.com/Guaranteer/VidSTG-Dataset","frameworks":[]}],"num_papers_in_archive":29,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/spatio-temporal-video-grounding-on-vidstg","task":"Spatio-Temporal Video Grounding","dataset_variant":"VidSTG","rows":3,"metrics":["Declarative m_vIoU","Declarative vIoU@0.3","Declarative vIoU@0.5","Interrogative m_vIoU","Interrogative vIoU@0.3","Interrogative vIoU@0.5"],"first_row_in_archive_order":{"model":"TA-STVG","paper":"/paper/knowing-your-target-target-aware-transformer","metrics":{"Declarative m_vIoU":"34.4","Declarative vIoU@0.3":"48.2","Declarative vIoU@0.5":"33.5","Interrogative m_vIoU":"29.5","Interrogative vIoU@0.3":"41.5","Interrogative vIoU@0.5":"28.0"},"code_links":[{"title":"HengLan/TA-STVG","url":"https://github.com/HengLan/TA-STVG"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/knowing-your-target-target-aware-transformer","title":"Knowing Your Target: Target-Aware Transformer Makes Better Spatio-Temporal Video Grounding","date":"2025-02-16","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/context-guided-spatio-temporal-video","title":"Context-Guided Spatio-Temporal Video Grounding","date":"2024-01-03","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":34,"samples_ran":21,"samples_unverified":13,"pointer_only_for_licence":34,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/tubedetr-spatio-temporal-video-grounding-with","title":"TubeDETR: Spatio-Temporal Video Grounding with Transformers","date":"2022-03-30","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"pointer_only_for_licence":0,"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":2,"samples_harvested":37,"samples_ran":24,"samples_unverified":13,"pointer_only_for_licence":34,"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."}