JayKarhade

Results 11 comments of JayKarhade

RuntimeError: CUDA out of memory.

@ShaoxiongYao did you find a fix or replacement for this issue?

Different results on two machines

Hi @WilliamKRobert @unlimblue did you guys find any fix to this solution? I built it from source as well however i face the same problem. For reference, i am running...

Can you share some implementation details about the result about 'K-Means Clustering of Frozen Diffusion Features'??

Did someone reproduce this?

Any plan for dinov2?

This is awesome! What would be a tentative release date for dinov2?

After running post_splatam_opt.py for densification, the newly generated npz file has worse performance compared to splatam.py

Hi @LemonSoda-RPG i got some time to check this - it seems to work fine on my end. Could you share your config files and post_splatam_opt.py file? I'd like to...

Question about SAM

Hi @Graysonggg we had done preliminary evaluations on SAM with those parameters, however we did not see better performance than DINOv2 and hence didn't probe further. The hyperparameters that have...

one error in AnLoc/demo/ anyloc_vlad_generate.py row of 120th

Hi @VClay , Thanks for pointing it out. It is indeed a string literal https://github.com/AnyLoc/AnyLoc/blob/2ae462dbce05a2741fe8d075a1313488804e29a3/demo/anyloc_vlad_generate.py#L62. Since it's already been defined, i don't think it should be generating an error ....

visualization

Hi @axelBarroso were there any updates on the code for visualization of the 3d point correspondences? It'd be really awesome to check this!

Question about training.

Hi @LiheYoung Is there a specific reason you do not update teacher model weights with Exponential Moving Averages of the student weights? Do you think this may improve performance further?

How to train from scratch or finetuning?

@yocabon i also tried the training script by using the entire Megadepth dataset with a larger batch size of 128 and for more epochs. While the loss goes down, the...