SCSampler: Sampling Salient Clips From Video for Efficient Action Recognition

@article{Korbar2019SCSamplerSS,
  title={SCSampler: Sampling Salient Clips From Video for Efficient Action Recognition},
  author={Bruno Korbar and Du Tran and L. Torresani},
  journal={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2019},
  pages={6231-6241}
}
  • Bruno Korbar, Du Tran, L. Torresani
  • Published 2019
  • Computer Science
  • 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
  • While many action recognition datasets consist of collections of brief, trimmed videos each containing a relevant action, videos in the real-world (e.g., on YouTube) exhibit very different properties: they are often several minutes long, where brief relevant clips are often interleaved with segments of extended duration containing little change. [...] Key Result Furthermore, we show that this yields significant gains in recognition accuracy compared to analysis of all clips or randomly/uniformly selected clips…Expand Abstract
    23 Citations
    Listen to Look: Action Recognition by Previewing Audio
    • 14
    • PDF
    FASTER Recurrent Networks for Efficient Video Classification
    • 4
    Large-Scale Weakly-Supervised Pre-Training for Video Action Recognition
    • 64
    • PDF
    Dynamic Sampling Networks for Efficient Action Recognition in Videos
    AR-Net: Adaptive Frame Resolution for Efficient Action Recognition
    • 2
    • Highly Influenced
    • PDF
    Shuffle and Attend: Video Domain Adaptation
    TimeGate: Conditional Gating of Segments in Long-range Activities
    • 2
    • Highly Influenced
    • PDF
    SAST: Learning Semantic Action-Aware Spatial-Temporal Features for Efficient Action Recognition
    Unified Multisensory Perception: Weakly-Supervised Audio-Visual Video Parsing
    • 1
    • PDF
    Pose And Joint-Aware Action Recognition

    References

    SHOWING 1-10 OF 65 REFERENCES
    Fast Temporal Activity Proposals for Efficient Detection of Human Actions in Untrimmed Videos
    • 166
    • PDF
    Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset
    • 1,933
    • Highly Influential
    • PDF
    Compressed Video Action Recognition
    • 118
    • Highly Influential
    • PDF
    Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
    • 1,537
    • PDF
    Long-Term Temporal Convolutions for Action Recognition
    • 526
    • PDF
    Learnable pooling with Context Gating for video classification
    • 179
    • PDF
    TURN TAP: Temporal Unit Regression Network for Temporal Action Proposals
    • 196
    • PDF
    Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs
    • 430
    • PDF
    Parsing Videos of Actions with Segmental Grammars
    • 138
    • PDF