YouTube-VOS: A Large-Scale Video Object Segmentation Benchmark
This dataset addresses a bottleneck for researchers in video analysis by enabling more robust training and evaluation of methods that capture spatial-temporal dependencies, though it is incremental as it builds on existing data collection efforts.
The authors tackled the lack of large-scale datasets for video object segmentation by creating YouTube-VOS, which contains 4,453 video clips and 94 object categories, establishing it as the largest dataset in this area and providing baselines for future algorithm development.
Learning long-term spatial-temporal features are critical for many video analysis tasks. However, existing video segmentation methods predominantly rely on static image segmentation techniques, and methods capturing temporal dependency for segmentation have to depend on pretrained optical flow models, leading to suboptimal solutions for the problem. End-to-end sequential learning to explore spatialtemporal features for video segmentation is largely limited by the scale of available video segmentation datasets, i.e., even the largest video segmentation dataset only contains 90 short video clips. To solve this problem, we build a new large-scale video object segmentation dataset called YouTube Video Object Segmentation dataset (YouTube-VOS). Our dataset contains 4,453 YouTube video clips and 94 object categories. This is by far the largest video object segmentation dataset to our knowledge and has been released at http://youtube-vos.org. We further evaluate several existing state-of-the-art video object segmentation algorithms on this dataset which aims to establish baselines for the development of new algorithms in the future.