Skip to content

posed_images

posed_images #

PosedImage — one photograph, the camera that took it, and the depth it measured.

PosedImage #

Bases: Struct

A frame a splat is trained on.

pose places the camera in the world with its intrinsics expressed at the photograph's own size, so every pixel is a ray; depth is the frame's captured depth map, Missing where none was measured. A dataset is an Array of these, and arrays of frames in one world concatenate.

Source code in capturegraph-lib/capturegraph/recipes/splats/posed_images.py 
13
14
15
16
17
18
19
20
21
22
23
24
25
class PosedImage(Struct):
    """A frame a splat is trained on.

    ``pose`` places the camera in the world with its intrinsics expressed at
    the photograph's own size, so every pixel is a ray; ``depth`` is the
    frame's captured depth map, ``Missing`` where none was measured. A
    dataset is an ``Array`` of these, and arrays of frames in one world
    concatenate.
    """

    image: Image
    pose: Pose
    depth: Depth