SAM_TensorRT

SAM_TensorRT copied to clipboard

how to get mobile_sam_ecoder/decoder.onnx

8

In [MobileSAM](https://github.com/ChaoningZhang/MobileSAM), there is only one weight file named mobile_sam.pt, how to get encoder and decoder weigt separately? Could you share the script for that?

busyyang

weiqwert

Does it support Nano SAM ?

2

Nice work! In the quest for faster speeds, is the framework applicable to Nano SAM ?

Rocky-YU

thanks your work

1

thanks for your contribution. is this repo still suitable for edgeSAM?

SoulProficiency

SAM HQ?

1

Hi, thank you for this code. Can it support SAM HQ? https://github.com/SysCV/sam-hq

antithing

Result not correct

1

Hi, thank you for your efforts. I am currently testing the code using the "dog.jpg" example, but the results are not accurate. Initially, I drew a bounding box around the...

sctrueew

forward error

1

init enegien sucess,but forward error: kERROR: 3: [executionContext.cpp::nvinfer1::rt::ShapeMachineContext::resolveSlots::2509] Error Code 3: API Usage Error (Parameter check failed at: executionContext.cpp::nvinfer1::rt::ShapeMachineContext::resolveSlots::2509, condition: allInputDimensionsSpecified(routine) )

machine52vision

TensorRT 8.6 change onnx to trt error

5

1) /usr/local/TensorRT-8.6.1.6//bin/trtexec --onnx=./models/mobile_sam_encoder.onnx --saveEngine=./models/mobile_sam_encoder.engine --memPoolSize=workspace:10240 2) TensorRT 8.6 change onnx to trt error log: &&&& RUNNING TensorRT.trtexec [TensorRT v8601] # /usr/local/TensorRT-8.6.1.6//bin/trtexec --onnx=./models/mobile_sam_encoder.onnx --saveEngine=./models/mobile_sam_encoder.engine --memPoolSize=workspace:10240 [11/03/2023-16:35:20] [I] === Model Options ===...

avBuffer

1024x1024

1

Does this algorithm process 1024x1024 ，no matter what size picture it processes ，Is the output image also that large?

xiangw369

RT，do you modify some code about mobiesam when you convert encoder.onnx？my converted encoder.onnx"s accuracy is not as good as you

m-wei