A CLIP-Enhanced Method for Video-Language Understanding
This work addresses video-text tasks for AI researchers, but it is incremental as it builds on existing CLIP and VALUE frameworks.
The paper tackled video-language understanding by proposing a CLIP-enhanced method to incorporate image-text pretrained knowledge, achieving a 2.4% improvement in Meta-Ave score from 57.58 to 60.00 on the VALUE benchmark.
This technical report summarizes our method for the Video-And-Language Understanding Evaluation (VALUE) challenge (https://value-benchmark.github.io/challenge\_2021.html). We propose a CLIP-Enhanced method to incorporate the image-text pretrained knowledge into downstream video-text tasks. Combined with several other improved designs, our method outperforms the state-of-the-art by $2.4\%$ ($57.58$ to $60.00$) Meta-Ave score on VALUE benchmark.