I managed to run the ResNetCorr model without any problem with the get_desp.py. However, when I try to run the SegStereo model, both SegStereo and KITTI fine tuned, I encountered a problem where it shows error: "Check failed: error == cudaSuccess (2 vs. 0) out of memory"
I am wondering if anyone else is facing this problem or have found a solution to this problem?
I think the reason I am facing this problem is that the model requires >12GB in GPU, as shown in the figure below:
and the GPU I am using is the GTX Titan X that has 12GB. But as shown in the KITTI website, the model was also deployed on a GTX Titan X card. So I do not understand why I am having this issue.
Operating System: Ubuntu 16.04
CUDA: 8
CuDNN: 5
It will be great if someone can help me to solve this problem!
Hi,
I managed to run the ResNetCorr model without any problem with the get_desp.py. However, when I try to run the SegStereo model, both SegStereo and KITTI fine tuned, I encountered a problem where it shows error: "Check failed: error == cudaSuccess (2 vs. 0) out of memory"
I am wondering if anyone else is facing this problem or have found a solution to this problem?
I think the reason I am facing this problem is that the model requires >12GB in GPU, as shown in the figure below: and the GPU I am using is the GTX Titan X that has 12GB. But as shown in the KITTI website, the model was also deployed on a GTX Titan X card. So I do not understand why I am having this issue.
Operating System: Ubuntu 16.04 CUDA: 8 CuDNN: 5
It will be great if someone can help me to solve this problem!
Thank you
The Caffemodel of ResNetCorr only requires 7GB: