MSU Video Alignment and Retrieval Benchmark Suite

Explore the best algorithms in different video alignment tasks

Key features of the Benchmark Suite

  • The most diverse dataset for alignment of near-duplicate videos:
    • 560 test pairs in each Benchmark with a total duration of ~2 million frames
    • Combinations of 13 frequent distortions obtained due to human/machine video editing and processing
  • Test your method on our Benchmarks:
    • Local time distortions: occur due to video processing (transmitting, compressing with low bitrate etc.)
    • Global time distortions: occur due to video editing
    • Mixed version for versatility testing on both distortion types
  • Find the best method for your near-duplicate video alignment requirements:
    • Each Benchmark contains three distortion presets representing use cases with different complexity

We appreciate new ideas. Please, write us an e-mail to aligners-benchmark@videoprocessing.ai

Leaderboard

The table below shows a comparison of Video alignment methods in Mixed time distortions benchmark using light distortions preset.

Charts

In these section you can see scatter plots with F1-score of alignment for each test pair in Mixed time distortions benchmark using light distortions preset. You can sort charts by the algorithm.

Full leaderboards and charts for each benchmark and preset you can see here.

Submit your algorithm

  1. Download input
    • Download the necessary folders for the benchmark you want to take part in here.
  2. Apply your algorithm
    • Align videos with your algorithm
  3. Send us result to aligners-benchmark@videoprocessing.ai

If you would like to participate read the full submission rules here.

Contacts

For questions and propositions, please contact us: aligners-benchmark@videoprocessing.ai

Learn more

Written on October 22, 2021