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

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

13 Dec 2022CVPR 2023 1arXiv:2212.06785archive 2025-07-28

Renrui Zhang, Liuhui Wang, Yu Qiao, Peng Gao, Hongsheng Li

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.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

The archive's official repository (zrrskywalker/i2p-mae) was not harvested; the 6 samples below come from 1 community repository.

Syntology 6 code samples harvested from 1 repository linked to this paper: 6 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 6 samples from 1 repository, 6 ran (2 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 6 samples that ran have no recorded date.

zrrskywalker/i2p-mae officialmentioned in papermentioned on GitHubpytorchSyntology: not harvested report
zrrskywalker/point-m2ae mentioned on GitHubpytorchMITSyntology: 6 samples, 6 ran (4 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

6 samples harvested: 6 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
2ran · our draft was wrong
2ran; the contract check could not run

Licence: 4 of the 6 samples are pointer only, meaning Syntology does not serve that copy's text. For 4 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from zrrskywalker/point-m2ae. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

zrrskywalker/point-m2ae community (archive-listed) 6 samples, 6 ran (4 instrument-failed) · 0 unverified
index_points in 49 papers models/modules.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · 449a0265144f6530 · report fixture: signature points:Tensor/3/float, idx:Tensor/2/int · returned Tensor · level recorded, date not recorded
farthest_point_sample in 64 papers datasets/ModelNetDataset.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · f80066a00e7156a2 · report fixture: signature point:ndarray/2/float, npoint:int/0/int · returned ndarray · level recorded 2026-09-25
load_modelnet_data in 8 papers datasets/ModelNetDataset.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · e609e218822ed309 · report fixture: signature config:Config/0/?, partition:str/0/str · returned tuple · level recorded 2026-09-24
pc_normalize in 81 papers
census: 2 implementations from 6 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
datasets/ModelNetDataset.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · 4783fbece52f500e · report fixture: signature pc:ndarray/2/float · returned ndarray · level recorded 2026-09-22
square_distance in 89 papers
census: 11 implementations from 18 papers, 4 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 3, kind float, dtype float32 · arg 2: rank 3, kind float, dtype float32
models/modules.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 3bfe172e686075cd · report fixture: signature src:Tensor/3/float, dst:Tensor/3/float · returned Tensor · level recorded, date not recorded
timeit in 59 papers segmentation/pointnet2_utils.py ran · honoured contract MIT (permissive) here; the stored copy is pointer only · b1227ddb721e2999 · report fixture: signature tag:str/0/str, t:float/0/float · returned float · level recorded 2026-09-26

Tasks

3D Point Cloud Classification3D Point Cloud Linear ClassificationDecoderFew-Shot 3D Point Cloud Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Point Cloud Classification ScanObjectNN I2P-MAE (no voting) OBJ-BG (OA) 94.15 #21 of 77 Archive leaderboard report
3D Point Cloud Classification ScanObjectNN I2P-MAE (no voting) OBJ-ONLY (OA) 91.57 #21 of 77 Archive leaderboard report
3D Point Cloud Classification ScanObjectNN I2P-MAE (no voting) Overall Accuracy 90.11 #21 of 77 Archive leaderboard report
3D Point Cloud Linear Classification ModelNet40 I2P-MAE Overall Accuracy 93.4 #3 of 20 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 10-way (10-shot) I2P-MAE Overall Accuracy 92.6 #15 of 31 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 10-way (10-shot) I2P-MAE Standard Deviation 5.0 #15 of 31 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 10-way (20-shot) I2P-MAE Overall Accuracy 95.5 #12 of 31 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 10-way (20-shot) I2P-MAE Standard Deviation 3.0 #12 of 31 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 5-way (10-shot) I2P-MAE Overall Accuracy 97.0 #11 of 30 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 5-way (10-shot) I2P-MAE Standard Deviation 1.8 #11 of 30 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 5-way (20-shot) I2P-MAE Overall Accuracy 98.3 #10 of 30 Archive leaderboard report
Few-Shot 3D Point Cloud Classification ModelNet40 5-way (20-shot) I2P-MAE Standard Deviation 1.3 #10 of 30 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

SVM

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections