zhubin

These instructions given in evaluation. md ,I tried every instruction python -m immatch.eval_hpatches --gpu 0 \ --config 'superpoint' 'ncnet' \ --task 'both' --save_npy \ --root_dir .

Konria version：0.6.5

For example, I replaced "attention" in "loftr" with "quardtree attention". Then I put the code in your code like "loftr". model = immatch.__dict__[class_name](args) loftr.py line23 self.model = LoFTR_(config=conf) KeyError: 'block_type'...

Ok, thank you. The problem has been solved. It is indeed a configuration problem. But the test results are all 0. What do you think might be the problem?

你们好，我想请教下问题，为什么我按照这个格式跑TUM数据集： print(i, t[0, 0], t[1, 0], t[2, 0], gt_pose[0, 3], gt_pose[1, 3], gt_pose[2, 3], file=log_fopen) img1 = keypoints_plot(img, vo) img2 = traj_plotter.update(t, gt_pose[:, 3]) img2只有一个点，t的值是0,最后生成的result是这样的： 0 0.0 0.0 0.0 1.686...