The video features used in TVC task

linjieli222 / HERO

Research code for EMNLP 2020 paper "HERO: Hierarchical Encoder for Video+Language Omni-representation Pre-training"

https://arxiv.org/abs/2005.00200

MIT License

230 stars 34 forks source link

The video features used in TVC task #5

Closed tuyunbin closed 3 years ago

tuyunbin commented 3 years ago

Hi, Lin jie, How can I repreduce the results in TVC task in your paper? Can you kindly provide the features you used in this task? Thanks.

linjieli222 commented 3 years ago

Sorry about the late reply and thanks for your interests.

The video features for TVC task are the same as the ones used for TVR and TVQA. @ChenRocks is working on releasing the finetuning code on TVC. Will let you know when it is out.

Thanks, Linjie

tuyunbin commented 3 years ago

Thank you for you reply and I'm looking forward to your released code.

linjieli222 commented 3 years ago

Hi @tuyunbin,

Code for TVC task has released. Thanks for your patience.

Best, Linjie

tuyunbin commented 3 years ago

Hi @linjieli222 , thank you very much!

tuyunbin commented 3 years ago

Hi, @linjieli222 , can you provide the results on TVC and TVR test-public results without pre-training?