heyoeyo
heyoeyo
The expected shape for box inputs is: Bx4 And if you're using points & labels, the shape should be BxNx2 and BxN respectively, where B is the batch size and...
There's some code on the [SAMv2 issue board](https://github.com/facebookresearch/segment-anything-2/issues/235#issuecomment-2312650295) that provides support for having multiple boxes for a single prompt. That code references changes to the newer code base, but the...
In general, if you have different shaped data, it would need to be processed in separate batches. In this case if you had multiple images with 2 bounding boxes you...
One thing to check is whether the `JSON.parse(fileEmbeddings)` part is completing successfully (I'm not sure if the error is occurring during/after that). It's possible that the embedding wasn't saved in...
Hi @CriusFission there may be a few things off here, but it's hard to say for sure. The main thing that stands out as strange is the size of your...
It should be possible to set the box input to `None` to disable it (i.e. prevent a box prompt from influencing the output). There's also a (learned) [not a point](https://github.com/facebookresearch/segment-anything/blob/6fdee8f2727f4506cfbbe553e23b895e27956588/segment_anything/modeling/prompt_encoder.py#L48)...
> a way to "selectively allow" prompting within batches As far as I can tell, the sparse embeddings have a shape of BxNx256, where N is related to the number...
The shape of the points should be `B x N x 2`, labels should have a shape of: `B x N` and boxes should have a shape of: `B x...
It should be possible to have 2 (or more) boxes, but they'll generate independent masks, since the '2' in this case will correspond to the batch part of the shape...
> Does that mean that there is no use in using these image encodings to compute similarity between two images? The embeddings are always going to be the same shape,...