PySceneDetect in Literature

PySceneDetect is a useful tool for statistical analysis of video. Below are links to various research articles/papers which have either used PySceneDetect as a part of their analysis, or propose more accurate detection algorithms using the current implementation as a comparison.

• Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers by Tsai-Shien Chen, Aliaksandr Siarohin, Willi Menapace, Ekaterina Deyneka, Hsiang-wei Chao, Byung Eun Jeon, Yuwei Fang, Hsin-Ying Lee, Jian Ren, Ming-Hsuan Yang, Sergey Tulyakov (2024)

• Stable Remaster: Bridging the Gap Between Old Content and New Displays by Nathan Paull, Shuvam Keshari, Yian Wong (2023)

• LoL-V2T: Large-Scale Esports Video Description Dataset by Tsunehiko Tanaka, Edgar Simo-Serra (2021)

• Online Detection of Action Start via Soft Computing for Smart City by Tian Wang, Yang Chen, Hongqiang Lv, Jing Teng, Hichem Snoussi, Fei Tao (2020)

• Thesis Project: Smart Shades and Cane for The Blind by Muhammad Hashim (2020)

• Movienet: a movie multilayer network model using visual and textual semantic cues by Youssef Mourchid, Benjamin Renoust, Olivier Roupin, Lê Văn, Hocine Cherifi & Mohammed El Hassouni (2019)

• NLP-Enriched Automatic Video Segmentation by Mohannad AlMousa, Rachid Benlamri, Richard Khoury (2018)

• Online Detection of Action Start in Untrimmed, Streaming Videos by Zheng Shou, Junting Pan, Jonathan Chan, Kazuyuki Miyazawa, Hassan Mansour, Anthony Vetro, Xavi Gir-i-Nieto, Shih-Fu Chang (2018)

• Story Understanding in Video Advertisements by Keren Ye, Kyle Buettner, Adriana Kovashka (2018)

This list is only provided for academic and research purposes, and is far from an exhaustive source of the uses of PySceneDetect in literature. If you think a particular submission is relevant and should be added to this list, feel free to raise an issue with your suggestion. Publicly available material is preferred, although not a requirement.

Scene Detection Methodology

You can find the source code for each scene detector in the scenedetect/detectors folder. Also see Issue #62: Reference of paper for the methods used on Github for a further discussion on detection methodologies. You are more than welcome to propose any new ideas on the issue tracker, or share a proof of concept using the Python API by creating a pull request.