SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2105.08756 · 2021

Pathdreamer: A World Model for Indoor Navigation

arXiv · PDF · Open in the Atlas

Code that ran

7 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 3 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2105.08756.json · badge: /api/ran/2105.08756.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
google-research/pathdreamer canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
FunctionStatusWhere it lives
cmap_to_label Ran
ran, violated the contract we drafted
google-research/pathdreamer/utils/utils.py
code served (permissive licence) · get_code("c76a434feac55807")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family int, shape (4, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small RGB image tensor (H=4, W=5, 3 channels for RGB)
image_tensor = np.random.randint(0, 256, size=(4, 5, 3), dtype=np.uint8)

# Create a colormap with 256 colors, each with 3 channels (RGB)
cmap = np.random.randint(0, 256, size=(256, 3), dtype=np.uint8)

ARGS = {
    "image_tensor": image_tensor,
    "cmap": cmap,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family int, shape (4, 5) — shape (4, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compute_kl Ran
ran, violated the contract we drafted
google-research/pathdreamer/models/point_cloud_models.py
code served (permissive licence) · get_code("5b5f0fe56d18c7b3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf

# Set seed for reproducibility
tf.random.set_seed(0)

# Create small tensors with distinct shapes to probe dimensions
batch_size = 2
dim = 4

# Create mean tensors
mu1 = tf.constant([[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]], dtype=tf.float32)
mu2 = tf.constant([[0.15, 0.25, 0.35, 0.45], [0.55, 0.65, 0.75, 0.85]], dtype=tf.float32)

# Create log-variance tensors
logvar1 = tf.constant([[-0.5, -0.4, -0.3, -0.2], [-0.1, 0.0, 0.1, 0.2]], dtype=tf.float32)
logvar2 = tf.constant([[-0.6, -0.5, -0.4, -0.3], [-0.2, -0.1, 0.0, 0.1]], dtype=tf.float32)

ARGS = {
    'mu1': mu1,
    'logvar1': logvar1,
    'mu2': mu2,
    'logvar2': logvar2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
equirectangular_pixel_rays Ran
ran, honoured the contract we drafted
google-research/pathdreamer/utils/pano_utils.py
code served (permissive licence) · get_code("b6af8e0d386da328")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (3, 32) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import math
import tensorflow as tf

# The function takes a single parameter: output_height
# which should be an integer representing the height of the equirectangular image

ARGS = {
    "output_height": 4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (3, 32) — shape (3, 32) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_intrinsic_matrix Ran
ran, honoured the contract we drafted
google-research/pathdreamer/utils/point_cloud_utils.py
code served (permissive licence) · get_code("59c1860ba9ec3e3c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (4, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import tensorflow as tf

# Define the argument for get_intrinsic_matrix
# The function takes a horizontal field of view (hfov) as a float parameter
# hfov should be a reasonable angle value in radians

hfov_value = 1.0471975511965976  # approximately 60 degrees in radians

ARGS = {
    "hfov": hfov_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (4, 4) — shape (4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_world_to_image_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
google-research/pathdreamer/utils/pano_utils.py
code served (permissive licence) · get_code("e20343bcc9bddd30")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf

# Set seed for reproducibility
tf.random.set_seed(0)

# Define arguments for get_world_to_image_transform
image_shape = [480, 640]  # height, width
fov = tf.constant([0.785, 1.047], dtype=tf.float32)  # vertical and horizontal FOV in radians (45 and 60 degrees)
camera_intrinsics = None  # Will be computed from FOV
rotations = tf.constant([0.1, 0.2], dtype=tf.float32)  # pitch and heading in radians
rotation_matrix = None  # Not used when rotations is provided

ARGS = {
    'image_shape': image_shape,
    'fov': fov,
    'camera_intrinsics': camera_intrinsics,
    'rotations': rotations,
    'rotation_matrix': rotation_matrix,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (3, 3) — shape (3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
project_to_feat Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
google-research/pathdreamer/utils/point_cloud_utils.py
code served (permissive licence) · get_code("d3f3f7995e119e5b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small test tensors
batch_size = 2
num_points = 4
height = 8
width = 10
channels = 3

# transformed_coords: shape (N, 4, M) of (x, y, z, 1) values
# Values should be in normalized space roughly [-1, 1] for x,y and positive for z
transformed_coords = tf.constant([
    [[0.5, -0.3, 0.2, 0.1],
     [0.2, 0.4, 0.3, -0.1],
     [-0.5, 0.1, 0.5, 0.0],
     [0.0, -0.2, 0.4, 0.2]],
    [[0.3, 0.1, 0.2, 0.0],
     [-0.2, 0.3, 0.4, 0.1],
     [0.1, -0.4, 0.3, 0.0],
     [-0.3, 0.2, 0.5, -0.1]]
], dtype=tf.float32)  # shape (2, 4, 4)

# feats: shape (N, M, C) - feature values with channels
feats = tf.constant([
    [[1.0, 2.0, 3.0],
     [2.0, 3.0, 4.0],
     [3.0, 4.0, 5.0],
     [4.0, 5.0, 6.0]],
    [[5.0, 6.0, 7.0],
     [6.0, 7.0, 8.0],
     [7.0, 8.0, 9.0],
     [8.0, 9.0, 10.0]]
], dtype=tf.float32)  # shape (2, 4, 3)

height = 8
width = 10
depth_scale = 1.0
input_void_class = 0.0
output_void_class = 0.0

ARGS = {
    'transformed_coords': transformed_coords,
    'feats': feats,
    'height': height,
    'width': width,
    'depth_scale': depth_scale,
    'input_void_class': input_void_class,
    'output_void_class': output_void_class
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
reparameterize Ran
ran, honoured the contract we drafted
google-research/pathdreamer/models/point_cloud_models.py
code served (permissive licence) · get_code("579dc1e53eab3741")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Set random seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small tensors for mu and logvar
# mu: mean vector, shape (batch_size, latent_dim)
mu = tf.constant([[0.1, 0.2, 0.3, 0.4],
                   [0.5, 0.6, 0.7, 0.8]], dtype=tf.float32)

# logvar: log variance vector, same shape as mu
logvar = tf.constant([[-1.0, -0.5, -0.2, 0.0],
                      [-0.8, -0.3, -0.1, 0.2]], dtype=tf.float32)

ARGS = {
    "mu": mu,
    "logvar": logvar
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

People navigating in unfamiliar buildings take advantage of myriad visual, spatial and semantic cues to efficiently achieve their navigation goals. Towards equipping computational agents with similar capabilities, we introduce Pathdreamer, a visual world model for agents navigating in novel indoor environments. Given one or more previous visual observations, Pathdreamer generates plausible high-resolution 360 visual observations (RGB, semantic segmentation and depth) for viewpoints that have not been visited, in buildings not seen during training. In regions of high uncertainty (e.g. predicting around corners, imagining the contents of an unseen room), Pathdreamer can predict diverse scenes, allowing an agent to sample multiple realistic outcomes for a given trajectory. We demonstrate that Pathdreamer encodes useful and accessible visual, spatial and semantic knowledge about human environments by using it in the downstream task of Vision-and-Language Navigation (VLN). Specifically, we show that planning ahead with Pathdreamer brings about half the benefit of looking ahead at actual observations from unobserved parts of the environment. We hope that Pathdreamer will help unlock model-based approaches to challenging embodied navigation tasks such as navigating to specified objects and VLN.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2105.08756")
get_code_for_paper("2105.08756")
have("2105.08756")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2105.08756.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2105.08756.svg)](https://syntology.ai/paper/2105.08756)

Connect an agent — have() is free.