yformer/EfficientSAM
EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything
Jupyter NotebookApache-2.0
Issues
- 1
what is the minimum memory required for the GPU
#41 opened by ghm666 - 1
When there are multiple points as prompts (for example, more than five points), some point areas cannot be segmented.
#65 opened by daeing - 0
Is it possible to fine-tune it to extract single-category objects based only on mask prompt
#70 opened by 1248493632 - 7
Is there GPU support for box-prompted SAM?
#58 opened by silinsi - 1
- 0
pre-trained parameters
#67 opened by Yy4441867 - 0
pre-trained parameters
#68 opened by Yy4441867 - 1
how to use background point
#64 opened by KingerFzb - 2
eval on COCO dataset
#38 opened by shawnricecake - 3
Saliency Segment CODE WANTED
#54 opened by happyfox-dot - 2
CoreML
#52 opened by FerusAndBeyond - 4
- 2
- 1
- 1
SqueezeSAM
#51 opened by jrhe - 1
How to do Saliency segmentation?
#53 opened by xinli2008 - 2
how to train our dataset?thanks for your answer
#55 opened by wyk-study - 3
SAMI Module and Training Codes?
#56 opened by BingoNate - 1
Segment Anything CPP Wrapper for macOS
#57 opened by ryouchinsa - 1
- 1
multibox-prompt inference
#62 opened by BeliverK - 1
What does input_labels mean?
#61 opened by nightwindnohen - 1
- 6
"segment-anything" is slow
#35 opened by GitChanyoung - 1
- 2
What does input label mean in code?
#39 opened by ghm666 - 5
- 3
torch.onnx.errors.UnsupportedOperatorError: Exporting the operator 'aten::tile' to ONNX opset version 17 is not supported.
#47 opened by xwhkkk - 0
- 2
- 0
- 3
different in jit to pt model
#42 opened by Whatsetsthisend - 5
- 3
- 2
Code about Cross-Attention Decoder
#33 opened by 123456789asdfjkl - 2
will release the pretraining code?
#31 opened by Yut-tuat - 3
Integration to segment_anything package?
#19 opened by oranshayer - 1
Regarding the performance of SqueezeSAM
#34 opened by xiankgx - 7
- 3
How to use a text prompt.
#26 opened by GitChanyoung - 2
Finetuning for downstream segmentation task
#32 opened by agraham9966 - 1
- 2
The output mask is 3*w*h. I need to take the masked channel as the result. Or should I merge these channels?
#28 opened by loorr - 0
The inference time significantly increases when using multi-point prompts.
#29 opened by EmptyCity1995 - 6
How do you generate masked tokens?
#21 opened by lianjinke - 5
OutOfMemoryError
#25 opened by ccy610 - 1
Datasets used in pretraining
#23 opened by Cbtor - 2
- 0
OutOfMemoryError
#20 opened by ccy610 - 1
grounded-efficient-sam demo support
#15 opened by rentainhe