Rex Cheng

> Hello, it is a pleasure to read this code, but when I want to train my own dataset, I find that "python download_training_dataset.py" this line of code needs this.py...

Thank you for providing the videos. To preface, all models make mistakes and Cutie is no exception. Most of the frames look fairly reasonable to me. There might be a...

Feel free to re-open if there are any updates.

It is for curriculum learning: first, from easy to hard cases, then anneal back to 5, which is closer to what is used during inference. Note that max jump sets...

It generalizes. It is not unlike how CNN generalizes to different resolutions and how LLM generalizes to different sequence lengths with relative position embeddings. Learning a robust appearance representation (as...

Which part of the demo does not work for you?

Thanks, and sorry for the late reply. I believe this is a dependency error coming from HQSAM -- I have added some import guards. Please let me know if it...

Most of the datasets we used during evaluation are multi-object. Multi-object inference itself is not an issue. Do you know if our pretrained weight works? How are you training your...

I suspect that there is a bug in your training. Multi-object training/inference is pretty much just multiple single-object inferences in parallel with some object aggregation at the end. It should...

Technically you can, though we have not implemented this behavior.