aelimame

Based on that sample from the Ego-4D dataset (https://ego4d-data.org/docs/data/imu/) the sample rate is 200Hz (5ms each time step). If only T=2000 works, this means they expect the clips to correspond...

> It seems that we are supposed to use repeated padding? > > [`PadIm2Video(pad_type="repeat", ntimes=2)`](https://github.com/fabawi/ImageBind-LoRA/blob/127d8c4a03221bb187a02f9c1076e97d7e4b6c31/models/imagebind_model.py#L150) But that's for the image to video transformation ([forward() method](https://github.com/fabawi/ImageBind-LoRA/blob/127d8c4a03221bb187a02f9c1076e97d7e4b6c31/models/multimodal_preprocessors.py#LL433C9-L433C40)). It seems to convert a...

> I agree - I am just making the conjecture that since we want image-IMU alignments for training, if this is the procedure for image padding, it could work for...