top of page
-
Improved video captioning based on Show-and-Tell model.
-
image tags
-
audio feature (MFCC)
-
spatial attention
-
GAN (novel)
-
-
Attended TREC Video Retrieval Evaluation (TRECVID) 2017 ‘Video to Text’ (VTT) competition held by NIST.
Exploring CNN-LSTM Architectures for Video Captioning
Project @ CMU 11775 Large-Scale Multimedia Analysis
Mar 2017 - May 2017
 |
|---|
 |
 |
 |
|---|
bottom of page