Skip to content

Latest commit

 

History

History
7 lines (5 loc) · 380 Bytes

README.md

File metadata and controls

7 lines (5 loc) · 380 Bytes

Combining deep features and activity context via conditional random fields to improve activity recognition

  1. To prepare data, run prepare_all.py
  2. To learn and infer with the model, run predict.py
  3. The dataset for training and testing the C3D model can be found at https://drive.google.com/open?id=1mEnLUrnElfbqmfmQWPaBK8V-JX8zZ_sV
  4. To visualize the results, run viz.py