Comparison of Scene Change Detection Algorithms for Videos

Reddy, Bindu; Jadhav, Anita

doi:10.1109/acct.2015.44

Cited by 9 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Scene detection algorithms segregate a film into scenes based on changes in the video frame sequences determined by color histogram differences and frame intensity/brightness. Popular tools, such as PySceneDetect, utilize both of these methods enabling adjustable sensitivity for scene change thresholds (Castellano, 2023;Gruzman & Kostenkova, 2014;Reddy & Jadhav, 2015). For film scholars, these tools can greatly ease data extraction needed for certain statistical analyses, such as finding the mean shot length (MSL) or the comparison of shot length frequency across different films (Redfern, 2023b;Salt, 1974).…”

Section: 2) Step 2: Extracting the Framesmentioning

confidence: 99%

Cutting the Frame: An In-Depth Look at the Hitchcock Computer Vision Dataset

Siddiqui

2024

Journal of Open Humanities Data

View full text Add to dashboard Cite

This paper presents a comprehensive dataset comprising annotations generated by the Google Vision API for approximately 105,000 frames extracted from 15 Alfred Hitchcock films. These annotations include information about object detection, facial recognition, web-entity analysis, and explicit content filtering. With potential applications in the digital humanities and film studies, this dataset enables researchers to not only explore and evaluate cinematic content but also the ways that it resurfaces in various cultural contexts online.The paper provides a detailed account of the dataset creation process, which involved the decryption of the DVD, frame extraction, and costs of annotations. Additionally, the paper outlines future research possibilities based on the dataset. These include statistical analysis of frame content and labels to identify patterns and trends, comparisons of different computer vision algorithms to assess their accuracy and effectiveness, and the utilization of bipartite networks to explore mise-en-scene in films.

show abstract

Section: 2) Step 2: Extracting the Framesmentioning

confidence: 99%

Cutting the Frame: An In-Depth Look at the Hitchcock Computer Vision Dataset

Siddiqui

2024

Journal of Open Humanities Data

View full text Add to dashboard Cite

show abstract

“…Вісник Київського національного університету імені Тараса Шевченка Серія фізико-математичні науки 2021, 1 Bulletin of Taras Shevchenko National University of Kyiv Series Physics & Mathematics періоду, а також усіх людей або речі, пов'язані з цим. Поняття сцени є корисним у багатьох випадках:…”

Section: методиunclassified

“…Деякі важливі результати щодо цієї задачі можна знайти у [1][2][3][4][5][6][7]. Здебільшого, сучасні роботи концентруються на певній підзадачі:…”

Section: методиunclassified

Scene Change Localization in a Video

Bieda

2021

BKNUPhM

View full text Add to dashboard Cite

Millions of videos are uploaded each day to Youtube and similar platforms. One of the many issues that these services face is the extraction of useful metadata. There are a lot of tasks that arise with the processing of videos. For example, putting an ad is better in the middle of a video, and as an advertiser, one would probably prefer to show the ad in between scene cuts, where it would be less intrusive. Another example is when one would like to watch only through the most interesting or important pieces of video recording. In many cases, it is better to have an automatic scene cut detection approach instead of manually labeling thousands of videos. The scene change detection can help to analyze video-stream automatically: which characters appear in which scenes, how they interact and for how long, their relations and importance, and also to track many other issues. The potential solution can rely on different facts: objects appearance, contrast or intensity changed, other colorization, background chang, and also sound changes. In this work, we propose the method for effective scene change detection, which is based on thresholding, and also fade-in/fade-out scene analysis. It uses computer vision and image analysis approaches to identify the scene cuts. Experiments demonstrate the effectiveness of the proposed scene change detection approach.

show abstract

“…4, for example, a wide range of genres exist, and each trailer is labelled with on average at least 3 genres, while the year of the trailers is diverse from 1930s to the present day. Processing of the Dataset: The dataset is pre-processed, with scene detection performed using the PyScene Detect [36], extracting individual clips from each trailer. While, we remove the first and last frames to mitigate poor scene detection.…”

Section: Mmx-trailer-20: Multi-model Experts Trailer Datasetmentioning

confidence: 99%

Rethinking movie genre classification with fine-grained semantic clustering

Fish¹,

Weinbren²,

Gilbert³

2020

Preprint

View full text Add to dashboard Cite

Movie genre classification is an active research area in machine learning. However, due to the limited labels available, there can be large semantic variations between movies within a single genre definition. We expand these 'coarse' genre labels by identifying 'fine-grained' semantic information within the multi-modal content of movies. By leveraging pre-trained 'expert' networks, we learn the influence of different combinations of modes for multi-label genre classification. Using a contrastive loss, we continue to fine-tune this 'coarse' genre classification network to identify high-level intertextual similarities between the films across all genre labels. This leads to a more 'fine-grained' and detailed clustering, based on semantic similarities while still retaining some genre information. Our approach is demonstrated on a newly introduced multi-modal 37,866,450 frame, 8,800 movie trailer dataset, MMX-Trailer-20, which includes precomputed audio, location, motion, and image embeddings.

show abstract

Comparison of Scene Change Detection Algorithms for Videos

Cited by 9 publications

References 2 publications

Cutting the Frame: An In-Depth Look at the Hitchcock Computer Vision Dataset

Cutting the Frame: An In-Depth Look at the Hitchcock Computer Vision Dataset

Scene Change Localization in a Video

Rethinking movie genre classification with fine-grained semantic clustering

Contact Info

Product

Resources

About