SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.06753 · 2021

Learning Open-World Object Proposals without Learning to Classify

arXiv · PDF · Open in the Atlas

Code that ran

2 of 4 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, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 2 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/2108.06753.json · badge: /api/ran/2108.06753.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
ibaiGorordo/ONNX-Object-Localization-Network community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
KostadinovShalon/oln-mask community (classifier label)
classifier basis not recorded on the link
0 ran · 2 no recorded run
FunctionStatusWhere it lives
get_peak_color Ran
ran on a synthesized input
ibaiGorordo/ONNX-Object-Localization-Network/oln/utils.py
code served (permissive licence) · get_code("9fbf566658156939")
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 1, dtype family int, shape (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 cv2
import numpy as np

# Create a small test image with shape (height, width, channels)
# Using a 10x10 image with 3 channels (BGR)
image = np.random.randint(0, 256, size=(10, 10, 3), dtype=np.uint8)

# Create a bounding box in format [x_min, y_min, x_max, y_max]
# Valid coordinates within the image bounds
bbox = [1, 1, 8, 8]

ARGS = {
    "image": image,
    "bbox": bbox
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (3,) — shape (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
utils_draw_detections Ran
ran on a synthesized input
ibaiGorordo/ONNX-Object-Localization-Network/oln/utils.py
code served (permissive licence) · get_code("0ab51b4ec7aba937")
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 (100, 80, 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 cv2
import numpy as np

# Create a small test image (RGB format, uint8)
# Use distinct dimensions: 100x80 to make it clear
image = np.random.randint(0, 256, (100, 80, 3), dtype=np.uint8)

# Create bounding boxes in format [x1, y1, x2, y2]
# Make sure they are within image bounds
detections = [
    [10, 15, 40, 50],    # bbox 1: x1=10, y1=15, x2=40, y2=50
    [45, 20, 75, 65]     # bbox 2: x1=45, y1=20, x2=75, y2=65
]

# Create corresponding confidence scores
scores = [0.85, 0.92]

# Boolean flag for whether to draw scores
draw_scores = True

# Alpha value for blending mask and detection image (0.0 to 1.0)
mask_alpha = 0.3

ARGS = {
    "image": image,
    "detections": detections,
    "scores": scores,
    "draw_scores": draw_scores,
    "mask_alpha": mask_alpha
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — 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 (100, 80, 3) — shape (100, 80, 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
mask_target Not yet run
unverified (harvested, no recorded run)
KostadinovShalon/oln-mask/mmdet/core/mask/mask_target.py
code served (permissive licence) · get_code("c5f7386f5eef1a7c")
How it ran

no invocation recorded.

mask_target_single Not yet run
unverified (harvested, no recorded run)
KostadinovShalon/oln-mask/mmdet/core/mask/mask_target.py
code served (permissive licence) · get_code("4a6d7b6f2759aa71")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 2 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
mask_target 1 other paper
1 later
2111.15552 (2021-11)
NeuSample: Neural Sample Field for Efficient View Synthesis
mask_target_single 1 other paper
1 later
2111.15552 (2021-11)
NeuSample: Neural Sample Field for Efficient View Synthesis

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

Object proposals have become an integral preprocessing steps of many vision pipelines including object detection, weakly supervised detection, object discovery, tracking, etc. Compared to the learning-free methods, learning-based proposals have become popular recently due to the growing interest in object detection. The common paradigm is to learn object proposals from data labeled with a set of object regions and their corresponding categories. However, this approach often struggles with novel objects in the open world that are absent in the training set. In this paper, we identify that the problem is that the binary classifiers in existing proposal methods tend to overfit to the training categories. Therefore, we propose a classification-free Object Localization Network (OLN) which estimates the objectness of each region purely by how well the location and shape of a region overlap with any ground-truth object (e.g., centerness and IoU). This simple strategy learns generalizable objectness and outperforms existing proposals on cross-category generalization on COCO, as well as cross-dataset evaluation on RoboNet, Object365, and EpicKitchens. Finally, we demonstrate the merit of OLN for long-tail object detection on large vocabulary dataset, LVIS, where we notice clear improvement in rare and common categories.

For agents

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

get_harvested_code_for_paper("2108.06753")
get_code_for_paper("2108.06753")
have("2108.06753")

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

curl https://syntology.ai/api/ran/2108.06753.json

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

[![Syntology run record](https://syntology.ai/api/ran/2108.06753.svg)](https://syntology.ai/paper/2108.06753)

Connect an agent — have() is free.