Fine-tuning LLaVA on prior TRECVid video-text pairs improves scores on the VTT24 captioning test set for BLEU, METEOR, CIDEr, and CIDEr-D, while SPICE and STS results are mixed.
Motion driven approaches to shot boundary detection, low-level feature extraction and bbc rushes characterization at TRECVid 2005,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
PolySmart @ TRECVid 2024 Video Captioning (VTT)
Fine-tuning LLaVA on prior TRECVid video-text pairs improves scores on the VTT24 captioning test set for BLEU, METEOR, CIDEr, and CIDEr-D, while SPICE and STS results are mixed.