SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.06785 · 2022

Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders

arXiv · PDF · Open in the Atlas

Code that ran

6 of 6 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 0 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2212.06785.json · badge: /api/ran/2212.06785.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
zrrskywalker/point-m2ae canonical (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
FunctionStatusWhere it lives
index_points Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zrrskywalker/point-m2ae/models/modules.py
code served (permissive licence) · get_code("449a0265144f6530")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 3, dtype family float, shape (2, 4, 3) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T23:32:40Z). 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 tensors with distinct dimensions
# points: [B, N, C] where B=2 (batch), N=5 (num points), C=3 (channels)
points = torch.randn(2, 5, 3)

# idx: [B, S] where B=2 (batch), S=4 (num samples to index)
# Values should be valid indices into the N=5 dimension
idx = torch.tensor([[0, 2, 1, 4], [3, 1, 0, 2]], dtype=torch.long)

ARGS = {
    "points": points,
    "idx": idx,
}
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 Tensor, ndim 3, dtype family float, shape (2, 4, 3) — shape (2, 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
farthest_point_sample Ran
ran; our contract draft was wrong, not the code
zrrskywalker/point-m2ae/datasets/ModelNetDataset.py
pointer only (licence: MIT); the stored copy get_code serves, from Yuzhe-Fu/FlashFPS, is pointer only (licence: NOASSERTION) · get_code("f80066a00e7156a2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from Yuzhe-Fu/FlashFPS, 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.

load_modelnet_data Ran
ran; our contract draft was wrong, not the code
zrrskywalker/point-m2ae/datasets/ModelNetDataset.py
pointer only (licence: MIT); the stored copy get_code serves, from EvenJoker/Point-PEFT, is pointer only (licence: NONE) · get_code("e609e218822ed309")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from EvenJoker/Point-PEFT, 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.

pc_normalize Ran
ran, honoured the contract we drafted
zrrskywalker/point-m2ae/datasets/ModelNetDataset.py
pointer only (licence: MIT); the stored copy get_code serves, from ajhamdi/MVTN, is pointer only (licence: NONE) · get_code("4783fbece52f500e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ajhamdi/MVTN, 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.

square_distance Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zrrskywalker/point-m2ae/models/modules.py
code served (permissive licence) · get_code("3bfe172e686075cd")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 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 torch

torch.manual_seed(0)

# Create source points: [B, N, C]
# B=2 (batch), N=3 (num source points), C=4 (coordinates)
src = torch.randn(2, 3, 4)

# Create destination points: [B, M, C]
# B=2 (batch), M=5 (num destination points), C=4 (coordinates)
dst = torch.randn(2, 5, 4)

ARGS = {
    "src": src,
    "dst": dst,
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 5) — shape (2, 3, 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
timeit Ran
ran, honoured the contract we drafted
zrrskywalker/point-m2ae/segmentation/pointnet2_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from duasoi/Mind2Cloud, is pointer only (licence: NONE) · get_code("b1227ddb721e2999")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from duasoi/Mind2Cloud, 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.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 2 the same month, 6 later) and 0 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
square_distance 75 other papers
23 dated earlier, 2 the same month, 49 later, 1 same year, order not determinable · 9 only as an assembly brick
1706.02413 (2017-06)
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2003.00410 (2020-03)
PF-Net: Point Fractal Network for 3D Point Cloud Completion
and 72 more not named here
pc_normalize 47 other papers
16 dated earlier, 31 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
1912.12098 (2019-12)
Quaternion Equivariant Capsule Networks for 3D Point Clouds
2003.00492 (2020-03)
PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
and 44 more not named here
farthest_point_sample 43 other papers
13 dated earlier, 30 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2003.00492 (2020-03)
PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling
2010.05272 (2020-10)
IF-Defense: 3D Adversarial Point Cloud Defense via Implicit Function based Restoration
and 40 more not named here
index_points 42 other papers
16 dated earlier, 26 later · 6 only as an assembly brick
1706.02413 (2017-06)
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2003.13479 (2020-03)
RPM-Net: Robust Point Matching using Learned Features
and 39 more not named here
timeit 33 other papers
8 dated earlier, 1 the same month, 24 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2003.00410 (2020-03)
PF-Net: Point Fractal Network for 3D Point Cloud Completion
2003.04618 (2020-03)
Convolutional Occupancy Networks
and 30 more not named here
load_modelnet_data 2 other papers
1 dated earlier, 1 later
2203.06604 (2022-03)
Masked Autoencoders for Point Cloud Self-supervised Learning
2404.07989 (2024-04)
Any2Point: Empowering Any-modality Large Models for Efficient 3D Understanding

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

Pre-training by numerous image data has become de-facto for robust 2D representations. In contrast, due to the expensive data acquisition and annotation, a paucity of large-scale 3D datasets severely hinders the learning for high-quality 3D features. In this paper, we propose an alternative to obtain superior 3D representations from 2D pre-trained models via Image-to-Point Masked Autoencoders, named as I2P-MAE. By self-supervised pre-training, we leverage the well learned 2D knowledge to guide 3D masked autoencoding, which reconstructs the masked point tokens with an encoder-decoder architecture. Specifically, we first utilize off-the-shelf 2D models to extract the multi-view visual features of the input point cloud, and then conduct two types of image-to-point learning schemes on top. For one, we introduce a 2D-guided masking strategy that maintains semantically important point tokens to be visible for the encoder. Compared to random masking, the network can better concentrate on significant 3D structures and recover the masked tokens from key spatial cues. For another, we enforce these visible tokens to reconstruct the corresponding multi-view 2D features after the decoder. This enables the network to effectively inherit high-level 2D semantics learned from rich image data for discriminative 3D modeling. Aided by our image-to-point pre-training, the frozen I2P-MAE, without any fine-tuning, achieves 93.4% accuracy for linear SVM on ModelNet40, competitive to the fully trained results of existing methods. By further fine-tuning on on ScanObjectNN's hardest split, I2P-MAE attains the state-of-the-art 90.11% accuracy, +3.68% to the second-best, demonstrating superior transferable capacity. Code will be available at https://github.com/ZrrSkywalker/I2P-MAE.

For agents

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

get_harvested_code_for_paper("2212.06785")
get_code_for_paper("2212.06785")
have("2212.06785")

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

curl https://syntology.ai/api/ran/2212.06785.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.06785.svg)](https://syntology.ai/paper/2212.06785)

Connect an agent — have() is free.