Numerous research efforts and progresses have been done to extract video key frames in recent years, but the existing approaches have high computational complexity, and do not capture the main visual content effectively. Traditional clustering algorithms, such as K-means, require some prior knowledge to determine the initial parameters, most of them need to be specified manually, and it would be a tough job to define the optimum parameter. Ĭlustering is a powerful technique for statistical data analysis, used in many fields, including machine learning, pattern recognition, image analysis, information retrieval, data compression, and computer graphics. Key-frame extraction has been recognized as one of the important research issues in video information retrieval. The use of key frames greatly reduces the quantity of data required in video browsing and provides an organizational framework for dealing with video content. Key frames provide a suitable abstraction and framework for video indexing, browsing, and retrieval. However, video processing is a relatively time-consuming task due to the large and unstructured format of video data. Among all the media types (text, image, graphic, audio, and video), video is the most expressive one because it combines all the other media information together. The user might not have always adequate time to watch the entire video and the integral video content might not be the interest or important for the user. Handling such volume of content becomes a challenge for the implementation of the real-time applications, such as video surveillance, educational purposes, video lectures, and sports highlights. IntroductionĪdvancements in digital storage, content distribution, and digital video recorders result in making the recording of the digital content procedure easy.
The density peak clustering algorithm (DPCA) model is shown to be more effective than the other four models in precision and F-measure. Finally, several algorithms are compared in the experiments. Then, a cluster validation technique, the silhouette coefficient, is employed to get the appropriate number of clusters without setting any clustering parameters.
Furthermore, by using the transformed one-dimensional eigenvector, the fixed number of features can be extracted for images with different sizes. The HSV histogram is used as color features for each frame, which reduces the amount of data. To address the problem, this study proposes a key-frame extraction method based on HSV (hue, saturation, value) histogram and adaptive clustering. Key-frame extraction is an important technique in video analysis, which provides an organizational framework for dealing with video content and reduces the amount of data required in video indexing. Some applications with high real-time requirements, such as object detection, require strong online video storage and analysis capabilities.
Use -vf yadif=1 as interlace filter to prevent interlacing artifacts: ffmpeg -i input.mpg -vf yadif=1 -c:v libx265 -crf 30 -c:a libopus -b:a 56k -frame_duration 60 output.Along with the fast development of digital information technology and the application of Internet, video data begins to grow explosively. For non-interlaced videos ffmpeg -i input.mpg -c:v libx265 -crf 30 -c:a libopus -b:a 56k -frame_duration 60 output.mkv For interlaced videos My recommendation is to use CRF 30 for lower-quality videos like analog grainy-ish videos, CRF 23 where you want to preserve the utmost quality, and CRF 26 for everything else. However, these are painfully slow, running at 0.0046x speed (libaom-av1) and 0.076x speed (libvpx-vp9) compared to 1.5x speed for libx265 for a test video since these encoders seem not to be highly optimized yet.ĬRF means constant rate factor, higher values mean better quality but larger file size. I also tried VP9 and AV1 which should result in even lower file size at the same quality.
#How to use ffmpeg to encode videos with hsv colorspace archive#
When you don’t want your video archive to eat up too much space, I recommend encoding them as H.265 and OPUS as these codecs provide excellent quality with typically less than half the bitrate of older formats.
0 kommentar(er)
