{"url":"/dataset/fsc147","name":"FSC147","full_name":null,"description_markdown":"We introduce a dataset of 147 object categories containing over 6000 images that are suitable for the few-shot counting task. We collected and annotated images ourselves. Our dataset consists of 6135 images across a di- verse set of 147 object categories, from kitchen utensils and office stationery to vehicles and animals. The object count in our dataset varies widely, from 7 to 3731 objects, with an average count of 56 objects per image. In each image, each object instance is annotated with a dot at its approxi- mate center. In addition, three object instances are selected randomly as exemplar instances; these exemplars are also annotated with axis-aligned bounding boxes.","description_withheld":null,"homepage":"https://github.com/cvlab-stonybrook/LearningToCountEverything","introduced_date":"2021-04-16","introduced_date_note":null,"introduced_by":{"paper":"/paper/learning-to-count-everything","title":"Learning To Count Everything","first_author":"Viresh Ranjan","url":null},"license":{"name":"MIT LICENSE","url":"https://github.com/cvlab-stonybrook/LearningToCountEverything/blob/master/LICENSE"},"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Object Counting","url":"/task/object-counting","datasets_with_task":"/datasets/task/object-counting"},{"name":"Training-free Object Counting","url":"/task/training-free-object-counting","datasets_with_task":"/datasets/task/training-free-object-counting"},{"name":"Few-shot Object Counting and Detection","url":"/task/few-shot-object-counting-and-detection","datasets_with_task":"/datasets/task/few-shot-object-counting-and-detection"},{"name":"Zero-Shot Counting","url":"/task/zero-shot-counting","datasets_with_task":"/datasets/task/zero-shot-counting"},{"name":"Exemplar-Free Counting","url":"/task/exemplar-free-counting","datasets_with_task":"/datasets/task/exemplar-free-counting"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["FSC147"],"data_loaders":[{"repo":"https://github.com/cvlab-stonybrook/LearningToCountEverything","url":"https://github.com/cvlab-stonybrook/LearningToCountEverything","frameworks":["pytorch"]}],"num_papers_in_archive":58,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/object-counting-on-fsc147","task":"Object Counting","dataset_variant":"FSC147","rows":19,"metrics":["MAE(test)","MAE(val)","RMSE(test)","RMSE(val)"],"first_row_in_archive_order":{"model":"CountGD","paper":"/paper/countgd-multi-modal-open-world-counting","metrics":{"MAE(test)":"5.74","MAE(val)":"7.1","RMSE(test)":"24.09","RMSE(val)":"26.08"},"code_links":[{"title":"niki-amini-naieni/CountGD","url":"https://github.com/niki-amini-naieni/CountGD"},{"title":"niki-amini-naieni/countx","url":"https://github.com/niki-amini-naieni/countx"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/exemplar-free-counting-on-fsc147","task":"Exemplar-Free Counting","dataset_variant":"FSC147","rows":9,"metrics":["MAE(test)","RMSE(test)","MAE(val)","RMSE(val)"],"first_row_in_archive_order":{"model":"SAVE","paper":"/paper/save-self-attention-on-visual-embedding-for","metrics":{"MAE(test)":"8.92","MAE(val)":"8.89","RMSE(test)":"80.39","RMSE(val)":"35.83"},"code_links":[{"title":"AhmedZgaren/Save","url":"https://github.com/AhmedZgaren/Save"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/few-shot-object-counting-and-detection-on","task":"Few-shot Object Counting and Detection","dataset_variant":"FSC147","rows":4,"metrics":["MAE(test)","RMSE(test)","AP50(test)","AP(test)"],"first_row_in_archive_order":{"model":"GeCo","paper":"/paper/a-novel-unified-architecture-for-low-shot","metrics":{"AP(test)":"43.42","AP50(test)":"75.06","MAE(test)":"7.91","RMSE(test)":"54.28"},"code_links":[{"title":"jerpelhan/GeCo","url":"https://github.com/jerpelhan/GeCo"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/training-free-object-counting-on-fsc147","task":"Training-free Object Counting","dataset_variant":"FSC147","rows":2,"metrics":["MAE"],"first_row_in_archive_order":{"model":"Omnicount","paper":"/paper/omnicount-multi-label-object-counting-with","metrics":{"MAE":"18.63"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/save-self-attention-on-visual-embedding-for","title":"SAVE: Self-Attention on Visual Embedding for Zero-Shot Generic Object Counting","date":"2025-02-10","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/a-novel-unified-architecture-for-low-shot","title":"A Novel Unified Architecture for Low-Shot Counting by Detection and Segmentation","date":"2024-09-27","rows_on_this_dataset":3,"code_links":1,"syntology":null},{"paper":"/paper/gca-sun-a-gated-context-aware-swin-unet-for","title":"GCA-SUNet: A Gated Context-Aware Swin-UNet for Exemplar-Free Counting","date":"2024-09-18","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/countgd-multi-modal-open-world-counting","title":"CountGD: Multi-Modal Open-World Counting","date":"2024-07-05","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":9,"samples_ran":4,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/learning-spatial-similarity-distribution-for","title":"Learning Spatial Similarity Distribution for Few-shot Object Counting","date":"2024-05-20","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":10,"samples_ran":4,"samples_unverified":6,"pointer_only_for_licence":10,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/dave-a-detect-and-verify-paradigm-for-low","title":"DAVE -- A Detect-and-Verify Paradigm for Low-Shot Counting","date":"2024-04-25","rows_on_this_dataset":3,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":18,"samples_ran":16,"samples_unverified":2,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/omnicount-multi-label-object-counting-with","title":"OmniCount: Multi-label Object Counting with Semantic-Geometric Priors","date":"2024-03-08","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/point-segment-and-count-a-generalized-1","title":"Point Segment and Count: A Generalized Framework for Object Counting","date":"2024-01-01","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/semantic-generative-augmentations-for-few","title":"Semantic Generative Augmentations for Few-Shot Counting","date":"2023-10-26","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/training-free-object-counting-with-prompts","title":"Training-free Object Counting with Prompts","date":"2023-06-30","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/open-world-text-specified-object-counting","title":"Open-world Text-specified Object Counting","date":"2023-06-02","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/vision-transformer-off-the-shelf-a-surprising","title":"Vision Transformer Off-the-Shelf: A Surprising Baseline for Few-Shot Class-Agnostic Counting","date":"2023-05-08","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/scale-prior-deformable-convolution-for","title":"Scale-Prior Deformable Convolution for Exemplar-Guided Class-Agnostic Counting","date":"2022-12-21","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/a-low-shot-object-counting-network-with","title":"A Low-Shot Object Counting Network With Iterative Prototype Adaptation","date":"2022-11-15","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":2,"samples_ran":0,"samples_unverified":2,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/countr-transformer-based-generalised-visual","title":"CounTR: Transformer-based Generalised Visual Counting","date":"2022-08-29","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":1,"samples_unverified":3,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/few-shot-object-counting-and-detection","title":"Few-shot Object Counting and Detection","date":"2022-07-22","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":0,"samples_unverified":4,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/learning-to-count-anything-reference-less","title":"Learning to Count Anything: Reference-less Class-agnostic Counting with Weak Supervision","date":"2022-05-20","rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":1,"samples_unverified":4,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/represent-compare-and-learn-a-similarity","title":"Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting","date":"2022-03-16","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":0,"samples_unverified":1,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/iterative-correlation-based-feature","title":"Few-shot Object Counting with Similarity-Aware Feature Enhancement","date":"2022-01-22","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/exemplar-free-class-incremental-learning-via","title":"Exemplar-free Class Incremental Learning via Discriminative and Comparable One-class Classifiers","date":"2022-01-05","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/object-counting-you-only-need-to-look-at-one","title":"Object Counting: You Only Need to Look at One","date":"2021-12-11","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/learning-to-count-everything","title":"Learning To Count Everything","date":"2021-04-16","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":0,"samples_unverified":1,"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":9,"samples_harvested":54,"samples_ran":26,"samples_unverified":28,"pointer_only_for_licence":10,"papers_with_no_sample_that_ran":4,"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."}