Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Questions about pose estimation #496

Open
own295554312 opened this issue Jun 23, 2024 · 2 comments
Open

Questions about pose estimation #496

own295554312 opened this issue Jun 23, 2024 · 2 comments

Comments

@own295554312
Copy link

The problem I'm wondering about is line 112 in evaluate_pose.py

for i in range(1, len(gt_global_poses)):
gt_local_poses.append(
np.linalg.inv(np.dot(np.linalg.inv(gt_global_poses[i - 1]), gt_global_poses[i])))

I know the point of this code is to get the pose from frame i-1 to frame i
nevertheless

np.dot(np.linalg.inv(gt_global_poses[i - 1]), gt_global_poses[i]))

This formula has already obtained the positional transformation from i-1 to i
Why do we need to find the inverse matrix again?

@own295554312
Copy link
Author

我想知道的问题是 evaluate_pose.py 中的第 112 行

for i in range(1, len(gt_global_poses)):
gt_local_poses.append(
np.linalg.inv(np.dot(np.linalg.inv(gt_global_poses[i - 1]), gt_global_poses[i])))

我知道这段代码的目的是获取从第 i-1 帧到第 i 帧的姿势
然而:

np.dot(np.linalg.inv(gt_global_poses[i - 1]), gt_global_poses[i]))

这个公式不是已经得到了从 i-1 到 i 的位姿变换嘛
为什么还要再次求逆矩阵?

@own295554312 own295554312 changed the title Questions about post estimation Questions about pose estimation Jun 23, 2024
@GLtangzero
Copy link

@own295554312
你好,我之前也是这里有问题,但论文里面明确说明的是,计算中间帧到前后两帧的T,因为算法是输入两帧图片,第一次输入是前一帧中间帧的顺序,因此计算出来的是前一帧中间帧的T,所以需要进行求逆矩阵获取中间帧前一帧的T。
而第二次的输入顺序则是中间帧后一帧,计算出来的T正好是我们需要的,因此不需要求逆矩阵。
你可以仔细看一下代码,只有在-1也就是第一类情况的时候才会计算逆矩阵。
个人愚见,希望有帮助!!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants