You're looking at a specific version of this model. Jump to the model overview.

vufinder /depth-anything-v3-metric-pano:085898c0

Playground API

Input schema

The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.

Field Type Default value Description
image
string
Input panoramic image, accepts JPG, JPEG, PNG, WEBP files
output_format
None
json
Output format of the prediction results.
return_depth
boolean
True
Whether to return depth image or not.
return_pcd
boolean
True
Whether to return a point cloud file or not.
return_views
boolean
False
Whether to return per view data. That includes image, metric depth, point maps, camera extrinsics and intrinsics.
processing_resolution
None
default
Maximum spatial dimension of input tensor passed to the model. The 'default' is 504 px, 36 patches 14 x 14 px. The 'custom' option allows to set the spatial dimension manually via the num_patches parameter.
num_patches
integer
36
The maximum spatial dimension of input tensor defined in number of patches. The patch dimension is 14 px. Only used if processing_resolution is set to 'custom'.
alpha_blend_onto
None
keep
Blend mode for images with alpha channels. The 'mean' mode blends the image onto ImageNet mean RGB values. The 'keep' mode keeps the original pixel values.
n_views
integer
24

Min: 12

Max: 64

Number of views to split the panorama image into.
view_fov
integer
90

Min: 40

Max: 90

Field of view of each view in degrees.
output_downscale
integer
4

Min: 1

Max: 8

Downscale factor for the output depth image. Low downscale factor improves quality at the cost of processing speed. Use higher values for higher resolution panoramas.
to_base64
boolean
True
Whether to return the arrays in JSON files as base64 strings with shape and dtype.

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema 
{'description': 'Class defining structure of the output.\n'
                '\n'
                'Attributes:\n'
                '    data (Path): Path to the NPZ/JSON file containing the '
                'panorama prediction results.\n'
                '    depth_image (Path | None): Optional path to depth image.\n'
                '    point_cloud (Path | None): Optional path to GLB point '
                'cloud.\n'
                '    views_data (list[Path] | None): Optional list of paths to '
                'the NPZ/JSON files containing\n'
                '        the prediction results for each view.',
 'properties': {'data': {'format': 'uri', 'title': 'Data', 'type': 'string'},
                'depth_image': {'format': 'uri',
                                'nullable': True,
                                'title': 'Depth Image',
                                'type': 'string'},
                'point_cloud': {'format': 'uri',
                                'nullable': True,
                                'title': 'Point Cloud',
                                'type': 'string'},
                'views_data': {'items': {'anyOf': [],
                                         'format': 'uri',
                                         'type': 'string'},
                               'nullable': True,
                               'title': 'Views Data',
                               'type': 'array'}},
 'required': ['data'],
 'title': 'PredictorOutput',
 'type': 'object'}