Jiashuo Yu

Speeding up the estimation procedure for higher cuda/torch version

3

Hi, I would like to provide an acceleration strategy that can address the problem of slow optical flow estimation speed in >pytorch1.1 version. Since we want to predict the dense...

Question about the onset feature extraction procedure

1

Hi, it seems that when extracting onset envelope, you use the audio percussive generated by the effects.hpss function as input of onset.onset_strength function instead of the raw spectrogram. I wonder...

Hi, Thanks for your great work! As you mentioned in the paper, the model is pretrained both on Kinetics-700 and AudioSet. Since only one pretrained model is available on the...

Any plan to release the video-SALMONN model?

1

Hi, thanks for your great work! I noticed your ICML paper video-SALMONN and it says the pretrained model and codes are released in this repository yet I cannot find them....