{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/large-scale-fast-and-accurate-shot-boundary","title":"Large-scale, Fast and Accurate Shot Boundary Detection through Spatio-temporal Convolutional Neural Networks","arxiv_id":"1705.03281","date":"2017-05-09","proceeding":null,"authors":["Ahmed Hassanien","Mohamed Elgharib","Ahmed Selim","Sung-Ho Bae","Mohamed Hefeeda","Wojciech Matusik"],"abstract":"Shot boundary detection (SBD) is an important pre-processing step for video\nmanipulation. Here, each segment of frames is classified as either sharp,\ngradual or no transition. Current SBD techniques analyze hand-crafted features\nand attempt to optimize both detection accuracy and processing speed. However,\nthe heavy computations of optical flow prevents this. To achieve this aim, we\npresent an SBD technique based on spatio-temporal Convolutional Neural Networks\n(CNN). Since current datasets are not large enough to train an accurate SBD\nCNN, we present a new dataset containing more than 3.5 million frames of sharp\nand gradual transitions. The transitions are generated synthetically using\nimage compositing models. Our dataset contain additional 70,000 frames of\nimportant hard-negative no transitions. We perform the largest evaluation to\ndate for one SBD algorithm, on real and synthetic data, containing more than\n4.85 million frames. In comparison to the state of the art, we outperform\ndissolve gradual detection, generate competitive performance for sharp\ndetections and produce significant improvement in wipes. In addition, we are up\nto 11 times faster than the state of the art.","url_abs":"http://arxiv.org/abs/1705.03281v2","url_pdf":"http://arxiv.org/pdf/1705.03281v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"large-scale-fast-and-accurate-shot-boundary","repo_url":"https://github.com/melgharib/DSBD","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"large-scale-fast-and-accurate-shot-boundary","repo_url":"https://github.com/Tangshitao/ClipShots","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"large-scale-fast-and-accurate-shot-boundary","repo_url":"https://github.com/soCzech/TransNetV2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"large-scale-fast-and-accurate-shot-boundary","repo_url":"https://github.com/wqliu657/TransNetV2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"boundary-detection","task_name":"Boundary Detection"},{"task_slug":"camera-shot-boundary-detection","task_name":"Camera shot boundary detection"},{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/camera-shot-boundary-detection-on-clipshots","task":"Camera shot boundary detection","dataset":"ClipShots","model":"DeepSBD","rank_in_archive_order":4,"of":4,"metrics":{"F1 score":"75.9"},"uses_additional_data":true},{"leaderboard":"/sota/camera-shot-boundary-detection-on-msu-shot","task":"Camera shot boundary detection","dataset":"MSU Shot Boundary Detection Benchmark","model":"PyScene-v2","rank_in_archive_order":3,"of":10,"metrics":{"F score":"0.7534","FPS":"86"},"uses_additional_data":false},{"leaderboard":"/sota/camera-shot-boundary-detection-on-msu-shot","task":"Camera shot boundary detection","dataset":"MSU Shot Boundary Detection Benchmark","model":"PyScene","rank_in_archive_order":7,"of":10,"metrics":{"F score":"0.7349","FPS":"86"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.03281","atlas_url":"https://app.syntology.ai/?focus=1705.03281","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}