跑自己图像出现问题

[jo-dean]

在跑sam的时候,出现下面这个问题

Grounded-Segment-Anything/segment_anything/segment_anything/modeling/mask_decoder.py", line 144, in predict_masks masks = (hyper_in @ upscaled_embedding.view(b, c, h * w)).view(b, -1, h, w) RuntimeError: cannot reshape tensor of 0 elements into shape [0, -1, 256, 256] because the unspecified dimension size -1 can be any value and is ambiguous

[917940234]

请问你使用python -m pip install -e GroundingDINO安装环境时，电脑自带C++相关环境吗？我现在不清楚我的错误是由其他问题引起的还是由C++环境版本引起的

[PoopBear1]

I encountered this issue as well

[jo-dean]

@917940234 对的，我之前装了gcc之类的。我跑有的图会出现这个问题

[zhang-jr]

It seems no prompt objects are segmented by sam in your images.

[jo-dean]

应该是的，已经可以了

[zhang-jr]

It seems no prompt objects are segmented by sam in your images.

If there are no output boxes generated by grounding dino, there is no need to apply sampredictor.

[PoopBear1]

应该是的，已经可以了

Could you kindly share how did you solve this issue? For my cases, this error especially raise when the "prompt-text" is not detected.

solved.

[Marcophono2]

@PoopBear1 Sharing your knowledge would have been fine.

[Dratlan]

If there is no output for grounding dino, this error occurs.