SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.00348 · CVPR · 2023

Modality-invariant Visual Odometry for Embodied Vision

Roman Bachmann, Amir Zamir, Marius Memmel

arXiv · PDF · Open in the Atlas

Code that ran

8 of 11 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; 0 honoured the contract we drafted, 3 violated it and 1 has no contract verdict; 4 failures of our instrument; 3 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/2305.00348.json · badge: /api/ran/2305.00348.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
memmelma/VO-Transformer canonical (classifier label)
classifier basis: census
8 ran · 3 no recorded run
FunctionStatusWhere it lives
build_2d_sincos_posemb Ran
ran on a synthesized input
memmelma/VO-Transformer/pointnav_vo/mmae/mmae_utils.py
code served (permissive licence) · get_code("95f1a074fba9a893")
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 Tensor, ndim 4, dtype family float, shape (1, 1024, 4, 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 torch
from einops import rearrange, repeat

torch.manual_seed(0)

ARGS = {
    "h": 4,
    "w": 3,
    "embed_dim": 1024,
    "temperature": 10000.0,
    "decode": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (1, 1024, 4, 3) — shape (1, 1024, 4, 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
has_file_allowed_extension Ran
ran, violated the contract we drafted
memmelma/VO-Transformer/pointnav_vo/model_utils/mmae/dataset_folder.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from dlyuangod/vit-1.58b, is pointer only (licence: NONE) · get_code("18ae3e6ef2cf02ec")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from dlyuangod/vit-1.58b, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

is_image_file Ran
ran, violated the contract we drafted
memmelma/VO-Transformer/pointnav_vo/model_utils/mmae/dataset_folder.py
code served (permissive licence) · get_code("1d15a850090b3c20")
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 bool, ndim 0, dtype family float, shape () 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 torch
import numpy as np
from typing import Tuple

torch.manual_seed(0)

# The function is_image_file takes a filename string and checks if it has an image extension
# It calls has_file_allowed_extension with the filename and IMG_EXTENSIONS tuple

ARGS = {
    "filename": "example_image.jpg"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () 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
make_dataset Ran
ran; our contract draft was wrong, not the code
memmelma/VO-Transformer/pointnav_vo/model_utils/mmae/dataset_folder.py
code served (permissive licence) · get_code("fb6aea1653b20bba")
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 list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 os
import os.path
import tempfile
from typing import Tuple, Dict, Optional, Callable, List

# Create a temporary directory structure for testing
temp_dir = tempfile.mkdtemp()

# Create class subdirectories and some dummy files
class_names = ["cat", "dog"]
for class_name in class_names:
    class_dir = os.path.join(temp_dir, class_name)
    os.makedirs(class_dir, exist_ok=True)
    # Create some dummy files with various extensions
    for i in range(2):
        with open(os.path.join(class_dir, f"image_{i}.jpg"), "w") as f:
            f.write("dummy")
        with open(os.path.join(class_dir, f"image_{i}.png"), "w") as f:
            f.write("dummy")
        with open(os.path.join(class_dir, f"other_{i}.txt"), "w") as f:
            f.write("dummy")

# Define the arguments
directory = temp_dir
class_to_idx = {"cat": 0, "dog": 1}
extensions = (".jpg", ".png")
is_valid_file = None

ARGS = {
    "directory": directory,
    "class_to_idx": class_to_idx,
    "extensions": extensions,
    "is_valid_file": is_valid_file,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family float — a list output has no array shape of its own; none is stated
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
make_fusion_block Ran
ran; our contract draft was wrong, not the code
memmelma/VO-Transformer/pointnav_vo/mmae/output_adapter_utils.py
code served (permissive licence) · get_code("5818d73b26500c83")
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 FeatureFusionBlock_custom. 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 torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "features": 64,
    "use_bn": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type FeatureFusionBlock_custom — a FeatureFusionBlock_custom output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a FeatureFusionBlock_custom output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not 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
make_scratch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
memmelma/VO-Transformer/pointnav_vo/mmae/output_adapter_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from nopqaq/test3r, is pointer only (licence: NOASSERTION) · get_code("45c2e73b37565105")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from nopqaq/test3r, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pair Ran
ran, violated the contract we drafted
memmelma/VO-Transformer/pointnav_vo/mmae/mmae_utils.py
code served (permissive licence) · get_code("6ba8cee9f5daea41")
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) and returned tuple, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 torch

torch.manual_seed(0)

# The function 'pair' takes a single parameter 't' and returns it as a tuple if it isn't already one
# We need to test both cases: when t is a tuple and when it's not

# Case 1: t is not a tuple (e.g., an integer, string, tensor, etc.)
t_not_tuple = 5

# Case 2: t is already a tuple
t_is_tuple = (3, 7)

# We'll use the non-tuple case as the primary test
ARGS = {
    "t": t_not_tuple
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family int — a tuple output has no array shape of its own; none is stated
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
trunc_normal_ Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
memmelma/VO-Transformer/pointnav_vo/mmae/mmae_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LHDuan/ConDaFormer, is pointer only (licence: NONE) · get_code("5436174f8c0e64e0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LHDuan/ConDaFormer, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pretrain_mmae_base Not yet run
unverified (harvested, no recorded run)
memmelma/VO-Transformer/pointnav_vo/mmae/mmae.py
code served (permissive licence) · get_code("2547dde00aeff3fd")
How it ran

no invocation recorded.

pretrain_mmae_large Not yet run
unverified (harvested, no recorded run)
memmelma/VO-Transformer/pointnav_vo/mmae/mmae.py
code served (permissive licence) · get_code("d6766d0a899eb434")
How it ran

no invocation recorded.

pretrain_mmae_tiny Not yet run
unverified (harvested, no recorded run)
memmelma/VO-Transformer/pointnav_vo/mmae/mmae.py
code served (permissive licence) · get_code("277cf7ad9089aa80")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 7 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 1 the same month, 6 later) and 4 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.

FunctionAlso present inEarliest-datedOthers named
trunc_normal_ 160 other papers
58 dated earlier, 1 the same month, 99 later, 2 same year, order not determinable · 36 only as an assembly brick
1911.10194 (2019-11)
Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation
2005.10242 (2020-05)
Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
2007.12911 (2020-07)
Tighter risk certificates for neural networks
and 157 more not named here
pair 90 other papers
30 dated earlier, 57 later, 2 same year, order not determinable, 1 undated · 8 only as an assembly brick
2004.13649 (2020-04)
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2103.15691 (2021-03)
ViViT: A Video Vision Transformer
and 87 more not named here
is_image_file 12 other papers
4 dated earlier, 7 later, 1 same year, order not determinable
2203.14415 (2022-03)
Mugs: A Multi-Granular Self-Supervised Learning Framework
2204.08381 (2022-04)
Multiple-environment Self-adaptive Network for Aerial-view Geo-localization
2205.03892 (2022-05)
ConvMAE: Masked Convolution Meets Masked Autoencoders
and 9 more not named here
has_file_allowed_extension 10 other papers
4 dated earlier, 5 later, 1 same year, order not determinable
2203.14415 (2022-03)
Mugs: A Multi-Granular Self-Supervised Learning Framework
2204.08381 (2022-04)
Multiple-environment Self-adaptive Network for Aerial-view Geo-localization
2205.03892 (2022-05)
ConvMAE: Masked Convolution Meets Masked Autoencoders
and 7 more not named here
make_dataset 6 other papers
2 dated earlier, 3 later, 1 same year, order not determinable
2205.03892 (2022-05)
ConvMAE: Masked Convolution Meets Masked Autoencoders
2208.07049 (2022-08)
Self-Supervised Vision Transformers for Malware Detection
aaai_25604 (2023 (year only))
One-for-All: Proposal Masked Cross-Class Anomaly Detection
and 3 more not named here
make_fusion_block 1 other paper
1 dated earlier
1907.10326 (2019-07)
From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation
make_scratch 1 other paper
1 later
2506.13750 (2025-06)
Test3R: Learning to Reconstruct 3D at Test Time

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

Effectively localizing an agent in a realistic, noisy setting is crucial for many embodied vision tasks. Visual Odometry (VO) is a practical substitute for unreliable GPS and compass sensors, especially in indoor environments. While SLAM-based methods show a solid performance without large data requirements, they are less flexible and robust w.r.t. to noise and changes in the sensor suite compared to learning-based approaches. Recent deep VO models, however, limit themselves to a fixed set of input modalities, e.g., RGB and depth, while training on millions of samples. When sensors fail, sensor suites change, or modalities are intentionally looped out due to available resources, e.g., power consumption, the models fail catastrophically. Furthermore, training these models from scratch is even more expensive without simulator access or suitable existing models that can be fine-tuned. While such scenarios get mostly ignored in simulation, they commonly hinder a model's reusability in real-world applications. We propose a Transformer-based modality-invariant VO approach that can deal with diverse or changing sensor suites of navigation agents. Our model outperforms previous methods while training on only a fraction of the data. We hope this method opens the door to a broader range of real-world applications that can benefit from flexible and learned VO models.

For agents

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

get_harvested_code_for_paper("2305.00348")
get_code_for_paper("2305.00348")
have("2305.00348")

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

curl https://syntology.ai/api/ran/2305.00348.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.00348.svg)](https://syntology.ai/paper/2305.00348)

Connect an agent — have() is free.