The goal of the Kinetics dataset is to help the computer vision and machine learning communities advance models for video understanding. Given this large human action classification dataset, it may be possible to learn powerful video representations that transfer to different video tasks.
Since I can't confirm the existence of the movie or director, the best approach is to inform the user that the reference isn't verified and offer alternative assistance. Also, remind them about legal ways to access content or provide historical context about 80s Philippine cinema if that's the direction they want.
Since George Estregan isn't a known figure in Philippine cinema, maybe the user made a mistake in the name. There's George Estregan, but more famously, there's the Esteban Esteban, but I'm not sure. Alternatively, maybe they're referring to George Estregan, a Filipino actor from the 80s? Let me verify.
Researching George Estregan: Not a prominent name. Maybe it's a mix-up with another name. Alternatively, the user could be referring to George Reinoso, but that's a stretch. Or perhaps Estela Estregan? Not sure.
Since I can't confirm the existence of the movie or director, the best approach is to inform the user that the reference isn't verified and offer alternative assistance. Also, remind them about legal ways to access content or provide historical context about 80s Philippine cinema if that's the direction they want.
Since George Estregan isn't a known figure in Philippine cinema, maybe the user made a mistake in the name. There's George Estregan, but more famously, there's the Esteban Esteban, but I'm not sure. Alternatively, maybe they're referring to George Estregan, a Filipino actor from the 80s? Let me verify.
Researching George Estregan: Not a prominent name. Maybe it's a mix-up with another name. Alternatively, the user could be referring to George Reinoso, but that's a stretch. Or perhaps Estela Estregan? Not sure.
1. Possible to use ImageNet checkpoints?
We allow finetuning from public ImageNet checkpoints for the supervised track -- but a link to the specific checkpoint should be provided with each submission.
2. Possible to use optical flow?
Flow can be used as long as not trained on external datasets, except if they are synthetic.
pinoy pene movies 80s sabik george estregan full repack
3. Can we train on test data without labels (e.g. transductive)?
No.
Since I can't confirm the existence of the
4. Can we use semantic class label information?
Yes, for the supervised track.
There's George Estregan, but more famously, there's the
5. Will there be special tracks for methods using fewer FLOPs / small models or just RGB vs RGB+Audio in the self-supervised track?
We will ask participants to provide the total number of model parameters and the modalities used and plan to create special mentions for those doing well in each setting, but not specific tracks.