sam-pt
sam-pt copied to clipboard
SAM-PT: Extending SAM to zero-shot video segmentation with point-based tracking.
Do you plan to release the code? if so, do you know when? Thanks, this work looks great!
Hi, Has this model been implemented without any training process? And is the only required training process the training of HQ-SAM independently? Alternatively, please provide the GPT/memory usage details for...
First of all thanks for this amazing work.. I am looking for a demo notebook/example to test it out on individual image/s, is there any available?
Thanks for your work, how much gpu and how much memory are required for training and testing?
the latest detectron2 requires pytorch>=1.18 but since the instructions tells us to do `conda install pytorch==1.12.0 torchvision==0.13.0 torchaudio==0.12.0 cudatoolkit=11.3 -c pytorch` it always fails. Installing detectron2 with a torch>=1.8 fails...
Hi, Can I ask if sam-pt can automatically track?
Hi, I need to create model outside sam-pt directory For now I get the following error `error locating target sam_pt.modelling.sam_pt.SamPt` How do I change `_target_` in config in order this...
我阅读了你们的论文,其中提到了通过提供初始帧掩码的方式进行采样,然后进行跟踪的方法。我想要知道你们提供了相关的测试代码吗?如果有提供就太好了 I read your paper which mentions a method of sampling by providing an initial frame mask and then tracking it. I would like to know if you have provided...
Hi again :) Your work is truly amazing, although, I wonder If model won't be able to work correctly with, for example, 3-5 frame video? For now, I have an...
Hi, thanks for your efforts and novel ideas. However, as I am running demo or experiments, I was asked to either enter API or API key not found. (3 options)...