Predicting daily activities from egocentric images using deep learning

@article{Castro2015PredictingDA,
  title={Predicting daily activities from egocentric images using deep learning},
  author={D. Castro and Steven Hickson and Vinay Bettadapura and E. Thomaz and Gregory D. Abowd and H. Christensen and Irfan Essa},
  journal={Proceedings of the 2015 ACM International Symposium on Wearable Computers},
  year={2015}
}
  • D. Castro, Steven Hickson, +4 authors Irfan Essa
  • Published 2015
  • Computer Science, Medicine
  • Proceedings of the 2015 ACM International Symposium on Wearable Computers
  • We present a method to analyze images taken from a passive egocentric wearable camera along with the contextual information, such as time and day of week, to learn and predict everyday activities of an individual. [...] Key Method Classification is conducted using a Convolutional Neural Network (CNN) with a classification method we introduce called a late fusion ensemble. This late fusion ensemble incorporates relevant contextual information and increases our classification accuracy. Our technique achieves an…Expand Abstract
    73 Citations

    Figures, Tables, and Topics from this paper

    Recognizing Activities of Daily Living from Egocentric Images
    • 13
    • Highly Influenced
    • PDF
    Batch-based activity recognition from egocentric photo-streams revisited
    • 8
    • Highly Influenced
    • PDF
    Batch-Based Activity Recognition from Egocentric Photo-Streams
    • 9
    • PDF
    Daily activity recognition from egocentric images
    • PDF
    Recognizing Daily Activities from Egocentric Photo-Streams
    • 1
    • Highly Influenced
    • PDF
    On the Role of Event Boundaries in Egocentric Activity Recognition from Photostreams
    • 2
    • PDF
    Activities of Daily Living Monitoring via a Wearable Camera: Toward Real-World Applications
    • Highly Influenced
    • PDF
    First Person Action Recognition Using Deep Learned Descriptors
    • 122
    • PDF
    Egocentric video description based on temporally-linked sequences
    • 15
    • PDF

    References

    ImageNet classification with deep convolutional neural networks
    • 58,240
    • Highly Influential
    • PDF