text
stringclasses 49
values |
---|
650 383 |
-1 |
167 |
171 |
162 |
162 |
-1 |
-1 |
163 |
163 |
164 |
163 |
164 |
163 |
-1 |
164 |
163 |
164 |
163 |
163 |
163 |
164 |
-1 |
167 |
164 |
166 |
163 |
-1 |
163 |
164 |
163 |
-1 |
164 |
165 |
165 |
163 |
164 |
164 |
166 |
163 |
167 |
164 |
164 |
163 |
163 |
169 |
-1 |
166 |
163 |
167 |
163 |
163 |
163 |
166 |
168 |
165 |
163 |
165 |
165 |
164 |
165 |
163 |
-1 |
164 |
164 |
179 |
166 |
-1 |
171 |
179 |
-1 |
163 |
167 |
168 |
166 |
164 |
165 |
-1 |
-1 |
166 |
164 |
164 |
162 |
164 |
167 |
167 |
-1 |
390 695 |
-1 |
198 |
200 |
201 |
197 |
-1 |
-1 |
200 |
199 |
197 |
200 |
197 |
Data Format
Here we explain the poses_bounds.npy
file format. This file stores a numpy array of size Nx17 (where N is the number of input videos). You can load the data using the following codes.
poses_arr = np.load(os.path.join(basedir, 'poses_bounds.npy'))
poses = poses_arr[:, :-2].reshape([-1, 3, 5]).transpose([1,2,0])
bds = poses_arr[:, -2:].transpose([1,0])
Each row of length 17 gets reshaped into a 3x5 pose matrix and 2 depth values that bound the closest and farthest scene content from that point of view.
The pose matrix is a 3x4 camera-to-world affine transform concatenated with a 3x1 column [image height, image width, focal length]
to represent the intrinsics (we assume the principal point is centered and that the focal length is the same for both x and y).
NOTE: In our dataset, the focal length for different cameras are different!!!
The right-handed coordinate system of the the rotation (first 3x3 block in the camera-to-world transform) is as follows: from the point of view of the camera, the three axes are [down, right, backwards]
which some people might consider to be [-y,x,z]
, where the camera is looking along -z
. (The more conventional frame [x,y,z]
is [right, up, backwards]
. The COLMAP frame is [right, down, forwards]
or [x,-y,-z]
.)
We also provide an example of our dataloader in dataloader.py
.
- Downloads last month
- 1,333