agenthong
agenthong
Hi, it's a great work! I follow the instruction to train and evaluate on cithscapes, while got following result which is slightly different from the paper.  Hence, how can...
``` [ 33%] Building NVCC (Device) object CMakeFiles/fusibile.dir/fusibile_generated_fusibile.cu.o /diskD/hzc/Self-Supervised-MVS/jdacs/fusion/fusibile/fusibile.cu(178): warning: variable "used_list" was set but never used /diskD/hzc/Self-Supervised-MVS/jdacs/fusion/fusibile/fusibile.cu(372): warning: variable "grid_size" was set but never used /diskD/hzc/Self-Supervised-MVS/jdacs/fusion/fusibile/fusibile.cu(375): warning: variable "block_size"...
I mean if you use 3 or 4 views in epipolar sample, like fuse features one more time by the 3rd view, you may get better results I guess. I'm...
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [1, 3, 3]] is at version 17; expected version 13 instead. Hint: enable...
Hi, @w2kun! Thanks for sharing this wonderful work! In the paper, it says scenes with moving objects are ignored. So how do you manually select no moving object scenes?
Hi, @huangyuan2020! I follow the instruction, but when I extract the main directions, it requires arguments --split. What else should I add? Thanks a lot!
Hi, @karfly Thanks for sharing this great repo. I've trained the model using human3.6 dataset. After that, I use 2D heatmaps of other images to unproject with my own carlibration...
It's a great repo. In your code there is a 'visualise_results' file, I run it but I only see this:  Is it correct and can I see the 3D...
Hi! Thanks for sharing such great work! I got a problem as follows:  Is there any mistake I make while preparing the NYU V2 dataset? Should I convert those...
Hi, @lauraset ! Thanks for sharing this repo! There are two pretrained models in `runs`, which are named `finetune.tar` and `V1.tar`. How did you train these two models? Could you...