SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.14841 · ECCV · 2024

MOD-UV: Learning Mobile Object Detectors from Unlabeled Videos

Bharath Hariharan, Yihong Sun

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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 5 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/2405.14841.json · badge: /api/ran/2405.14841.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
YihongSun/MOD-UV canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
FunctionStatusWhere it lives
cleanout_masks Ran
ran; our contract draft was wrong, not the code
YihongSun/MOD-UV/moduv/pseudo_labels/save_agg_pred.py
code served (permissive licence) · get_code("3d511cfd2d0f8dd1")
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 1, dtype family float, shape (2,) 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

torch.manual_seed(0)

# Create small mask tensors
# The function expects masks of shape (batch, height, width) or (batch, channels, height, width)
# We'll test with 4D input (batch, channels, height, width)
batch_size = 2
channels = 1
height = 4
width = 3

# Create binary masks with some overlap
masks_4d = torch.zeros(batch_size, channels, height, width, dtype=torch.float32)
masks_4d[0, 0, :2, :2] = 1.0  # First mask covers top-left
masks_4d[1, 0, 1:3, 1:3] = 1.0  # Second mask has some overlap with first

# Test parameters
descending = True  # Sort in descending order of mask size
filt_thrd = 0.3    # Filter threshold for overlap fraction

ARGS = {
    "masks": masks_4d,
    "descending": descending,
    "filt_thrd": filt_thrd,
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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_iou_matrix Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
YihongSun/MOD-UV/moduv/pseudo_labels/save_agg_pred.py
code served (permissive licence) · get_code("32f710aababf50d6")
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 2, dtype family float, shape (2, 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
import numpy as np

torch.manual_seed(0)

# Amasks: binary masks tensor of shape (batch_A, height, width)
# The function sums over the last two dimensions, so we need at least 3D tensors
Amasks = torch.randint(0, 2, (2, 4, 3), dtype=torch.float32)

# Bmasks: binary masks tensor of shape (batch_B, height, width)
# Same height and width as Amasks, but different batch size for variety
Bmasks = torch.randint(0, 2, (3, 4, 3), dtype=torch.float32)

ARGS = {
    "Amasks": Amasks,
    "Bmasks": Bmasks,
}
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 Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 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
filter_mask Ran
ran, violated the contract we drafted
YihongSun/MOD-UV/moduv/pseudo_labels/save_agg_pred.py
code served (permissive licence) · get_code("002e7c74d4ea4cfb")
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 1, dtype family float, shape (2,) 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

torch.manual_seed(0)

# masks: a 3D tensor of shape (batch, height, width) containing binary masks
# The function applies logical operations on the last two dimensions
masks = torch.randint(0, 2, (2, 4, 3), dtype=torch.bool)

# ref_bin_mask: a 2D tensor of shape (height, width) containing a reference binary mask
# It's broadcasted against all masks in the batch
ref_bin_mask = torch.randint(0, 2, (4, 3), dtype=torch.bool)

# filt_thrd: a scalar threshold for filtering
filt_thrd = 0.5

ARGS = {
    "masks": masks,
    "ref_bin_mask": ref_bin_mask,
    "filt_thrd": filt_thrd,
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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_img_path Ran
ran; our contract draft was wrong, not the code
YihongSun/MOD-UV/moduv/datasets/dataset.py
code served (permissive licence) · get_code("b557d65d7ad2f09e")
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 str. 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 os.path as osp

ARGS = {
    'data_dir': '/tmp/data',
    'segment_name': 'segment_001',
    'frame_idx': 42
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
read_img Ran
ran, honoured the contract we drafted
YihongSun/MOD-UV/moduv/utils.py
code served (permissive licence) · get_code("0d5bc119fb18b186")
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 Image. 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 tempfile
import os
from PIL import Image

# Create a temporary image file
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_image.png")

# Create a simple RGB image and save it
test_image = Image.new('RGB', (4, 3), color=(255, 0, 0))
test_image.save(temp_image_path)

ARGS = {
    'fname': temp_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
readlines Ran
ran; our contract draft was wrong, not the code
YihongSun/MOD-UV/moduv/utils.py
code served (permissive licence) · get_code("1ec3d12de0e96f1c")
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 1, 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 tempfile
import os

# Create a temporary text file with some content
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("line 1\nline 2\nline 3\n")
temp_file.close()

filename = temp_file.name

ARGS = {
    "filename": filename
}
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 1, 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
vis_2d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
YihongSun/MOD-UV/moduv/utils.py
code served (permissive licence) · get_code("c78b60f424caaa35")
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 3, dtype family int, shape (4, 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 torch
import numpy as np

torch.manual_seed(0)

# Create a small score map tensor of shape [1, 4, 3] (batch=1, height=4, width=3)
score_map_tensor = torch.randn(1, 4, 3)

# Alternative: create as numpy array
score_map_np = np.random.randn(1, 4, 3)

ARGS = {
    'score_map': score_map_tensor,
    'cmap': 'plasma',
    'vminmax': None,
    'max_perc': 95
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 3, 3) — shape (4, 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

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, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 6 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
readlines 7 other papers
6 dated earlier, 1 later
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
Zhou_R-MSFM_Recurrent_Multi-Scale_Feature_Modulation_for_Monocular_Depth_Estimating_ICCV_2021_paper (2021 (year only))
R-MSFM: Recurrent Multi-Scale Feature Modulation for Monocular Depth Estimating
2204.03636 (2022-04)
SurroundDepth: Entangling Surrounding Views for Self-Supervised Multi-Camera Depth Estimation
and 4 more not named here

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

Embodied agents must detect and localize objects of interest, e.g. traffic participants for self-driving cars. Supervision in the form of bounding boxes for this task is extremely expensive. As such, prior work has looked at unsupervised instance detection and segmentation, but in the absence of annotated boxes, it is unclear how pixels must be grouped into objects and which objects are of interest. This results in over-/undersegmentation and irrelevant objects. Inspired by human visual system and practical applications, we posit that the key missing cue for unsupervised detection is motion: objects of interest are typically mobile objects that frequently move and their motions can specify separate instances. In this paper, we propose MOD-UV, a Mobile Object Detector learned from Unlabeled Videos only. We begin with instance pseudolabels derived from motion segmentation, but introduce a novel training paradigm to progressively discover small objects and static-but-mobile objects that are missed by motion segmentation. As a result, though only learned from unlabeled videos, MOD-UV can detect and segment mobile objects from a single static image. Empirically, we achieve state-ofthe-art performance in unsupervised mobile object detection on Waymo Open, nuScenes, and KITTI Datasets without using any external data or supervised models. Code is available at github.com/YihongSun/MOD-UV.

For agents

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

get_harvested_code_for_paper("2405.14841")
get_code_for_paper("2405.14841")
have("2405.14841")

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

curl https://syntology.ai/api/ran/2405.14841.json

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

[![Syntology run record](https://syntology.ai/api/ran/2405.14841.svg)](https://syntology.ai/paper/2405.14841)

Connect an agent — have() is free.