linwk20
linwk20
Hi, Thanks for your quick reply, I will try running the latest code. Besides, I wonder is there other built-in debug way? For example, can I visualize the diverging process...
Hi. Thanks for your detailed reply! The result did not failed in my test, but is worse than A-LOAM, and that is why I post the issue. The evaluation metric...
Indeed, my dataset is collected in an underground indoor parking lot. Thanks, I will try tuning.
I have tried the OS-128 config and tuned keypoints num and voxel size. The ICP converged, but the result seems not become better. Do you have any advice for tuning?...
Cool, waiting for it.
I also found this problem, did you try atomicAdd(&gaussian_count[collected_id[j]], 1);? Will it be very slow?
waiting...
Maybe mini-batch optimization + gradient accumulation will help? currently the code using full-batch so is not scalable.
no, but it is easy to try, let me know if it works, you can try this on official small dataset first to see if accuracy degrade.
There are plenty of such MVS, one I know that runs at real time is Spann3r (https://hengyiwang.github.io/projects/spanner), but it is more like an academic paper, not trained on large dataset.