LoFTR
LoFTR copied to clipboard
Code for "LoFTR: Detector-Free Local Feature Matching with Transformers", CVPR 2021, T-PAMI 2022
Could you please share **the process of generating the dataset indices** of Megadepth, and explain **the specific meaning of the parameters in the .npz files** ? I am trying to...
Hi authors, Thank you for your repo, i want trainning your model with my custom dataset, i have some question? 1. What is ground truth of your model. i see...
Hey, thanks for your great work on this project! I found that the images in the scannet dataset are simply resized, but their intrinsics are not scaled (refer to SuperGlue),...
There seems to be some error when downloading the weights: ``` FileNotFoundError Traceback (most recent call last) [](https://localhost:8080/#) in 6 matcher.load_state_dict(torch.load("weights/indoor_ds.ckpt")['state_dict']) 7 elif image_type == 'outdoor': ----> 8 matcher.load_state_dict(torch.load("weights/outdoor_ds.ckpt")['state_dict']) 9...
Thanks for your great work! But I have a question: If I want to train a model for image matching, can I create my custom dataset using RGB image pairs...
尊敬的作者,您好。想请教一下如果我只有RGB图像,无法获得深度图信息,是不是没办法用这个模型进行训练
Hi, I have a question that how to train a model for HPatches? Do you train on it or train the model on MegaDepth or ScanNet, then evaluate the model...
In the snippet marked under the link: https://github.com/zju3dv/LoFTR/blob/c0546af290fe87aaa75f754afe2759e0b00412ea/src/utils/plotting.py#L78C5-L80C67, why are the indices at the end of the line `[[1, 0]]`? If I understand the code correctly, assuming a specific `b_id`,...