MART (ae-test split) - Appearance + Flow

Reported on 3 benchmarks across 1 task · 1 paper · 3 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision3 results

Video CaptioningonActivityNet Captions
BLEU4· 2020-05-11
10.33
best: 14.7 (VideoCoCa)
SOTA
MART: Memory-Augmented Recurrent Transformer for Coherent Video Paragraph Captioning arXiv:2005.05402
Video CaptioningonActivityNet Captions
CIDEr· 2020-05-11
23.42
best: 39.3 (VideoCoCa)
SOTA
MART: Memory-Augmented Recurrent Transformer for Coherent Video Paragraph Captioning arXiv:2005.05402
Video CaptioningonActivityNet Captions
METEOR· 2020-05-11
15.68
best: 17.97 (VLTinT (ae-test split) C3D/Ling)
SOTA
MART: Memory-Augmented Recurrent Transformer for Coherent Video Paragraph Captioning arXiv:2005.05402