{"url":"/task/scene-change-detection","name":"Scene Change Detection","slug":"scene-change-detection","description_markdown":"Scene change detection (SCD) refers to the task  of localizing changes and identifying change-categories given two scenes. A scene can be either an RGB (+D) image or a 3D reconstruction (point cloud).  If the scene is an image, SCD is a form of pixel-level prediction because each pixel in the image is classified according to a category. On the other hand, if the scene is point cloud, SCD is a form of point-level prediction because each point in the cloud is classified according to a category. \r\n\r\nSome example benchmarks for this task are VL-CMU-CD, PCD, and CD2014. Recently, more complicated benchmarks such as ChangeSim, HDMap, and Mallscape are released.\r\n\r\nModels are usually evaluated with the Mean Intersection-Over-Union (Mean IoU), Pixel Accuracy, or F1 metrics.","categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":25,"papers_with_code":13,"benchmarks":6,"benchmark_tables_in_archive":6,"benchmark_tables_shown":6,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":6,"subtasks":0,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/scene-change-detection-on-changesim","slug":"scene-change-detection-on-changesim","dataset":"ChangeSim","dataset_url":"/dataset/changesim","rows_in_archive":3,"metrics":["Category mIoU","macro F1"],"first_row_in_archive_order":{"model":"C-3PO","paper_title":"How to Reduce Change Detection to Semantic Segmentation","paper_url":"/paper/how-to-reduce-change-detection-to-semantic","paper_date":"2022-06-15","arxiv_id":"2206.07557","code_links":[{"title":"DoctorKey/C-3PO","url":"https://github.com/DoctorKey/C-3PO"}],"syntology":{"n":10,"n_ran":1,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/scene-change-detection-on-pcd","slug":"scene-change-detection-on-pcd","dataset":"PCD","dataset_url":"/dataset/pcd","rows_in_archive":2,"metrics":["F1-score"],"first_row_in_archive_order":{"model":"C-3PO (VGG-16)","paper_title":"How to Reduce Change Detection to Semantic Segmentation","paper_url":"/paper/how-to-reduce-change-detection-to-semantic","paper_date":"2022-06-15","arxiv_id":"2206.07557","code_links":[{"title":"DoctorKey/C-3PO","url":"https://github.com/DoctorKey/C-3PO"}],"syntology":{"n":10,"n_ran":1,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/scene-change-detection-on-unaligned-vl-cmu-cd","slug":"scene-change-detection-on-unaligned-vl-cmu-cd","dataset":"Unaligned-VL-CMU-CD (neighbor distance 2)","dataset_url":"/dataset/unaligned-vl-cmu-cd","rows_in_archive":2,"metrics":["F1-score"],"first_row_in_archive_order":{"model":"Robust-Scene-Change-Detection (Diff-View Augmentation)","paper_title":"Robust Scene Change Detection Using Visual Foundation Models and Cross-Attention Mechanisms","paper_url":"/paper/robust-scene-change-detection-using-visual","paper_date":"2024-09-25","arxiv_id":"2409.16850","code_links":[{"title":"ChadLin9596/Robust-Scene-Change-Detection","url":"https://github.com/ChadLin9596/Robust-Scene-Change-Detection"}],"syntology":null}},{"leaderboard":"/sota/scene-change-detection-on-vl-cmu-cd","slug":"scene-change-detection-on-vl-cmu-cd","dataset":"VL-CMU-CD","dataset_url":null,"rows_in_archive":2,"metrics":["F1-score"],"first_row_in_archive_order":{"model":"C-3PO (VGG-16)","paper_title":"How to Reduce Change Detection to Semantic Segmentation","paper_url":"/paper/how-to-reduce-change-detection-to-semantic","paper_date":"2022-06-15","arxiv_id":"2206.07557","code_links":[{"title":"DoctorKey/C-3PO","url":"https://github.com/DoctorKey/C-3PO"}],"syntology":{"n":10,"n_ran":1,"n_unverified":9,"n_pointer_only":0}}},{"leaderboard":"/sota/scene-change-detection-on-changevpr","slug":"scene-change-detection-on-changevpr","dataset":"ChangeVPR","dataset_url":"/dataset/changevpr","rows_in_archive":1,"metrics":["F1 score"],"first_row_in_archive_order":{"model":"GeSCF (zero-shot)","paper_title":"Towards Generalizable Scene Change Detection","paper_url":"/paper/towards-generalizable-scene-change-detection","paper_date":"2024-09-10","arxiv_id":"2409.06214","code_links":[{"title":"1124jaewookim/towards-generalizable-scene-change-detection","url":"https://github.com/1124jaewookim/towards-generalizable-scene-change-detection"}],"syntology":null}},{"leaderboard":"/sota/scene-change-detection-on-semantic-scene","slug":"scene-change-detection-on-semantic-scene","dataset":"Semantic Scene Understanding Challenge (passive actuation & ground-truth localisation)","dataset_url":null,"rows_in_archive":1,"metrics":["OMQ","avg_pairwise","avg_label","avg_spatial","avg_state_quality","avg_fp_quality"],"first_row_in_archive_order":{"model":"Demo_SCD","paper_title":null,"paper_url":null,"paper_date":"","arxiv_id":null,"code_links":[],"syntology":null}}],"datasets":[{"url":"/dataset/3r-scan","name":"3RScan","full_name":"","num_papers_in_archive":44},{"url":"/dataset/changesim","name":"ChangeSim","full_name":"ChangeSim","num_papers_in_archive":7},{"url":"/dataset/msu-shot-boundary-detection-benchmark","name":"MSU Shot Boundary Detection Benchmark","full_name":"","num_papers_in_archive":5},{"url":"/dataset/pcd","name":"PCD","full_name":"Poem Comprehensive Dataset","num_papers_in_archive":3},{"url":"/dataset/changevpr","name":"ChangeVPR","full_name":"","num_papers_in_archive":1},{"url":"/dataset/unaligned-vl-cmu-cd","name":"Unaligned-VL-CMU-CD (neighbor distance 2)","full_name":"","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":13,"of":13,"tagged_in_all":25,"items":[{"url":"/paper/learning-to-measure-change-fully","title":"Learning to Measure Change: Fully Convolutional Siamese Metric Networks for Scene Change Detection","date":"2018-10-22","arxiv_id":"1810.09111","repositories_listed":2,"syntology":null},{"url":"/paper/emplace-self-supervised-urban-scene-change","title":"EMPLACE: Self-Supervised Urban Scene Change Detection","date":"2025-03-22","arxiv_id":"2503.17716","repositories_listed":1,"syntology":{"n":7,"n_ran":4,"n_unverified":3,"n_pointer_only":7}},{"url":"/paper/robust-scene-change-detection-using-visual","title":"Robust Scene Change Detection Using Visual Foundation Models and Cross-Attention Mechanisms","date":"2024-09-25","arxiv_id":"2409.16850","repositories_listed":1,"syntology":null},{"url":"/paper/towards-generalizable-scene-change-detection","title":"Towards Generalizable Scene Change Detection","date":"2024-09-10","arxiv_id":"2409.06214","repositories_listed":1,"syntology":null},{"url":"/paper/zero-shot-scene-change-detection","title":"Zero-Shot Scene Change Detection","date":"2024-06-17","arxiv_id":"2406.11210","repositories_listed":1,"syntology":null},{"url":"/paper/unsupervised-change-detection-for-space","title":"Unsupervised Change Detection for Space Habitats Using 3D Point Clouds","date":"2023-12-04","arxiv_id":"2312.02396","repositories_listed":1,"syntology":null},{"url":"/paper/saras-net-scale-and-relation-aware-siamese","title":"SARAS-Net: Scale and Relation Aware Siamese Network for Change Detection","date":"2022-12-02","arxiv_id":"2212.01287","repositories_listed":1,"syntology":null},{"url":"/paper/differencing-based-self-supervised","title":"Differencing based Self-supervised pretraining for Scene Change Detection","date":"2022-08-11","arxiv_id":"2208.05838","repositories_listed":1,"syntology":null},{"url":"/paper/how-to-reduce-change-detection-to-semantic","title":"How to Reduce Change Detection to Semantic Segmentation","date":"2022-06-15","arxiv_id":"2206.07557","repositories_listed":1,"syntology":{"n":10,"n_ran":1,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/changesim-towards-end-to-end-online-scene","title":"ChangeSim: Towards End-to-End Online Scene Change Detection in Industrial Indoor Environments","date":"2021-03-09","arxiv_id":"2103.05368","repositories_listed":1,"syntology":null},{"url":"/paper/dr-tanet-dynamic-receptive-temporal-attention","title":"DR-TANet: Dynamic Receptive Temporal Attention Network for Street Scene Change Detection","date":"2021-03-01","arxiv_id":"2103.00879","repositories_listed":1,"syntology":null},{"url":"/paper/multi-temporal-scene-classification-and-scene","title":"Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion","date":"2020-06-03","arxiv_id":"2006.02176","repositories_listed":1,"syntology":null},{"url":"/paper/weakly-supervised-silhouette-based-semantic","title":"Weakly Supervised Silhouette-based Semantic Scene Change Detection","date":"2018-11-29","arxiv_id":"1811.11985","repositories_listed":1,"syntology":null}],"syntology_records":2,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}