CREStereo "next" onnx models for tensorrt
llschloesser opened this issue · 1 comments
llschloesser commented
Issue Type
Support
OS
Ubuntu
OS architecture
x86_64
Programming Language
C++, Python
Framework
ONNX, TensorRT
Model name and Weights/Checkpoints URL
crestereo_combined_iter20_480x640.onnx
Description
I am testing the CREStereo models at different resolutions, iters, combined vs. init, fp16, etc. with TensorRT. It appears that you or Ibai Gorordo created "next" models with a "flow" input. Can you make these available for download? Does that model imply that a traditional stereo algorithm such as SGBM can be used to seed the CREStereo network?
Relevant Log Output
No response
URL or source code for simple inference testing code
No response
PINTO0309 commented
All you do is isolate the model.
sne4onnx \
--input_onnx_file_path crestereo_combined_iter20_480x640.onnx \
--input_op_names next_left next_right init_output \
--output_op_names next_output \
--output_onnx_file_path crestereo_combined_iter20_480x640_next.onnx
