Discovering Video Clusters from Visual Features and Noisy Tags

  title={Discovering Video Clusters from Visual Features and Noisy Tags},
  author={Arash Vahdat and Guang-Tong Zhou and Greg Mori},
The Flip MMC framework proposed in this paper jointly optimizes the model parameters that describe each cluster, finds the best assignment of videos to clusters, and refines the tag labeling to reduce the noise in tag annotation. Similar to MMC, the Flip MMC optimization is a challenging non-convex optimization problem due to the discrete optimization that assigns videos to clusters and refines tag labels. Here this non-convex optimization problem is rewritten in unconstrained format as: 

From This Paper

Topics from this paper.
4 Citations
23 References
Similar Papers


Publications referenced by this paper.

Similar Papers

Loading similar papers…