omniglue

omniglue copied to clipboard

Question: inference takes long time

2

Thank you for providing the source code for this interesting work. However, I have a question regarding the inference time. On my device ( RTX 3090 (24GB)), a single inference...

lyp-deeplearning

Does your method require training?

1

I tried your method on my data and found that the matching was unsatisfactory, so I was wondering if your method requires training. Because I don't see the training code.

GZ-YourZY

Inference takes large GPU memory

6

Hi, Thanks for your great work. However, when I run your demo on an RTX 3090 (24GB) it takes about 22GB of GPU memory. Is this normal? Compared to LoFTR,...

plusgrey

Check failed: work_element_count > 0

2

Hello, I tried to run the demo code with torch=2.1.2 and tensorflow=2.12.0 with cuda=11.8. It shows the error below and the program aborted: > 2024-05-23 16:32:35.814947: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow...

qiaomu-miao

Hi, I was wondering (since the eval scripts havent been released) if you can give more specifics on how you calculate Acc@5°/ 10°/ 20° for the GSO, NAVI, and ScanNet...

jake-austin

Hello, will you upload the training code later?

4

wzj-888

Hello, thank you for your hard work! I suggest a minor code correction. 1. Improved GPU selection Currently, many people lack GPU memory or have problems requiring too much GPU...

Eunchan24

Can you open source the code for matcher inference, I'd like to see exactly how the matching is achieved and how to combine the feature descriptors of the dino

13559323996

I noticed that the 'descriptors0_dino' network input in omniglue has an embed_dim of 768, which is the size of dinov2_vitb14. Does this imply that the 384 embed_dim size of dinov2_vits14...

passinme

OutOfMemoryError

2

Why does the model consume so much memory? I followed your suggestions to use CUDA acceleration, but during the demo run, my 6GB VRAM was insufficient, resulting in a torch.cuda.OutOfMemoryError....

lyxxxyl