Annotation Cleaning for the MSR-Video to Text Dataset

We found that MSR-VTT dataset contains a lot of noisy annotations. After analyzing the data carefully, we put some efforts on cleaning the annotations. We retrained some models on the cleaned dataset and found experimental results improved compared to the previous models.

Requirements

1. Python 3
2. Jupyter Notebook
3. TensorFlow 1.13

Information

clean_process is the folder for cleaning MSR-VTT dataset. msrvtt_model is the folder for training a new model.

Links

GoogleDrive
Paper Link