A Simple and Powerful Global Optimization for Unsupervised Video Object Segmentation
This provides a simpler, scalable solution for video analysis tasks, though it is incremental as it matches rather than surpasses existing methods.
The paper tackles unsupervised video object segmentation by proposing a simple objective function that uses image features and optical flows to identify the main salient object, achieving performance on par with state-of-the-art methods on benchmarks like DAVIS2016, SegTrack-v2, and FBMS59.
We propose a simple, yet powerful approach for unsupervised object segmentation in videos. We introduce an objective function whose minimum represents the mask of the main salient object over the input sequence. It only relies on independent image features and optical flows, which can be obtained using off-the-shelf self-supervised methods. It scales with the length of the sequence with no need for superpixels or sparsification, and it generalizes to different datasets without any specific training. This objective function can actually be derived from a form of spectral clustering applied to the entire video. Our method achieves on-par performance with the state of the art on standard benchmarks (DAVIS2016, SegTrack-v2, FBMS59), while being conceptually and practically much simpler. Code is available at https://ponimatkin.github.io/ssl-vos.