Corpus ID: 49553994

A flexible model for training action localization with varying levels of supervision

@inproceedings{Chron2018AFM,
  title={A flexible model for training action localization with varying levels of supervision},
  author={Guilhem Ch{\'e}ron and Jean-Baptiste Alayrac and I. Laptev and C. Schmid},
  booktitle={NeurIPS},
  year={2018}
}
  • Guilhem Chéron, Jean-Baptiste Alayrac, +1 author C. Schmid
  • Published in NeurIPS 2018
  • Computer Science
  • Spatio-temporal action detection in videos is typically addressed in a fully-supervised setup with manual annotation of training videos required at every frame. [...] Key Method We investigate applications of such a model to training setups with alternative supervisory signals ranging from video-level class labels over temporal points or sparse action bounding boxes to the full per-frame annotation of action bounding boxes. Experiments on the challenging UCF101-24 and DALY datasets demonstrate competitive…Expand Abstract
    15 Citations

    Figures, Tables, and Topics from this paper

    Discovering Multi-Label Actor-Action Association in a Weakly Supervised Setting
    • PDF
    Uncertainty-Aware Weakly Supervised Action Detection from Untrimmed Videos
    • Highly Influenced
    • PDF
    SF-Net: Single-Frame Supervision for Temporal Action Localization
    • 5
    • PDF
    Action Recognition From Single Timestamp Supervision in Untrimmed Videos
    • 7
    • Highly Influenced
    • PDF
    End-to-End Learning of Visual Representations From Uncurated Instructional Videos
    • 56
    • PDF
    Anomaly Locality in Video Surveillance
    • 4
    • PDF
    Equivalent Classification Mapping for Weakly Supervised Temporal Action Localization
    • PDF
    y 1 ? ? ? ? ? open jar scoop sugar ? ? ? ? ? ? ?
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 49 REFERENCES
    Spot On: Action Localization from Pointly-Supervised Proposals
    • 93
    • Highly Influential
    • PDF
    Connectionist Temporal Modeling for Weakly Supervised Action Labeling
    • 137
    • PDF
    Automatic annotation of human actions in video
    • 270
    • PDF
    Weakly Supervised Action Learning with RNN Based Fine-to-Coarse Modeling
    • 90
    • PDF
    Human Action Localization with Sparse Spatial Supervision
    • 38
    Weakly Supervised Action Detection
    • 50
    • PDF
    Action Tubelet Detector for Spatio-Temporal Action Localization
    • 175
    • PDF
    Learning to Track for Spatio-Temporal Action Localization
    • 261
    • PDF
    Deep Learning for Detecting Multiple Space-Time Action Tubes in Videos
    • 151
    • PDF