Papers › Delivering Arbitrary-Modal Semantic Segmentation

Delivering Arbitrary-Modal Semantic Segmentation

2 Mar 2023CVPR 2023 1arXiv:2303.01480archive 2025-07-28

Jiaming Zhang, Ruiping Liu, Hao Shi, Kailun Yang, Simon Reiß, Kunyu Peng, Haodong Fu, Kaiwei Wang, Rainer Stiefelhagen

Multimodal fusion can make semantic segmentation more robust. However, fusing an arbitrary number of modalities remains underexplored. To delve into this problem, we create the DeLiVER arbitrary-modal segmentation benchmark, covering Depth, LiDAR, multiple Views, Events, and RGB. Aside from this, we provide this dataset in four severe weather conditions as well as five sensor failure cases to exploit modal complementarity and resolve partial outages. To make this possible, we present the arbitrary cross-modal segmentation model CMNeXt. It encompasses a Self-Query Hub (SQ-Hub) designed to extract effective information from any modality for subsequent fusion with the RGB representation and adds only negligible amounts of parameters (~0.01M) per additional modality. On top, to efficiently and flexibly harvest discriminative cues from the auxiliary modalities, we introduce the simple Parallel Pooling Mixer (PPX). With extensive experiments on a total of six benchmarks, our CMNeXt achieves state-of-the-art performance on the DeLiVER, KITTI-360, MFNet, NYU Depth V2, UrbanLF, and MCubeS datasets, allowing to scale from 1 to 81 modalities. On the freshly collected DeLiVER, the quad-modal CMNeXt reaches up to 66.30% in mIoU with a +9.10% gain as compared to the mono-modal baseline. The DeLiVER dataset and our code are at: https://jamycheung.github.io/DELIVER.html.

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

Syntology 7 code samples harvested from 1 repository linked to this paper: 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 7 samples from 1 repository, 7 ran (5 with no instrument failure; 2 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 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

jamycheung/DELIVER officialmentioned on GitHubpytorchApache-2.0Syntology: 7 samples, 7 ran (2 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

7 samples harvested: 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 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 · our draft was wrong
5ran

Licence: 6 of the 7 samples are pointer only, meaning Syntology does not serve that copy's text. For 6 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 jamycheung/DELIVER. “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.

jamycheung/DELIVER official repository 7 samples, 7 ran (2 instrument-failed) · 0 unverified
get_loss in 107 papers
census: 5 implementations from 5 papers, 5 distinct outputs across 4 buckets, one shared input per bucket
semseg/losses.py ran Apache-2.0 (permissive) · 3029c24a500aea0e · report fixture: signature loss_fn_name:str/0/str, ignore_label:int/0/int, cls_weights:NoneType/0/? · returned Dice · level recorded 2026-09-25
get_optimizer in 228 papers semseg/optimizers.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 624e13fbf125c2a5 · report fixture: signature model:callable/0/?, optimizer:str/0/str, lr:float/0/float, weight_decay:float/0/float · returned AdamW · level recorded 2026-09-24
get_scheduler in 122 papers semseg/schedulers.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 1190c7468e2cdaa7 · report fixture: signature scheduler_name:str/0/str, optimizer:SGD/0/?, max_iter:int/0/int, power:int/0/int, warmup_iter:int/0/int, warmup_ratio:float/0/float · returned WarmupPolyLR · level recorded 2026-09-24
get_train_augmentation semseg/augmentations.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 36844e88d38a0881 · report fixture: signature size:tuple/1/int, seg_fill:int/0/int · returned Compose · level recorded 2026-09-24
get_train_augmentation semseg/augmentations_mm.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 1ce37e1fa8008660 · report fixture: signature size:tuple/1/int, seg_fill:int/0/int · returned Compose · level recorded 2026-09-24
get_val_augmentation semseg/augmentations.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · c63b165d09644f24 · report fixture: signature size:list/1/int · returned Compose · level recorded 2026-09-24
get_val_augmentation semseg/augmentations_mm.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · 1096caea85dc06d3 · report fixture: signature size:tuple/1/int · returned Compose · level recorded 2026-09-24

Tasks

SegmentationSemantic SegmentationThermal Image Segmentation

Datasets

Introduced by this paper, per the archive.

DELIVER

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semantic Segmentation BJRoad CMNeXt IoU 63.22 #1 of 11 Archive leaderboard report
Semantic Segmentation DDD17 CMNeXt mIoU 72.67 #2 of 9 Archive leaderboard report
Semantic Segmentation DELIVER CMNeXt (RGB-D-E-LiDAR) mIoU 66.30 #3 of 9 Archive leaderboard report
Semantic Segmentation DELIVER CMNeXt (RGB-D-LiDAR) mIoU 65.50 #4 of 9 Archive leaderboard report
Semantic Segmentation DELIVER CMNeXt (RGB-D-Event) mIoU 64.44 #5 of 9 Archive leaderboard report
Semantic Segmentation DELIVER CMNeXt (RGB-Depth) mIoU 63.58 #6 of 9 Archive leaderboard report
Semantic Segmentation DELIVER CMNeXt (RGB-LiDAR) mIoU 58.04 #7 of 9 Archive leaderboard report
Semantic Segmentation DELIVER CMNeXt (RGB-Event) mIoU 57.48 #8 of 9 Archive leaderboard report
Semantic Segmentation DSEC CMNeXt mIoU 72.54 #2 of 9 Archive leaderboard report
Semantic Segmentation DeLiVER CMNeXt (RGB-D-E-LiDAR) mIoU 66.30 #5 of 26 Archive leaderboard report
Semantic Segmentation KITTI-360 CMNeXt (RGB-D-E-LiDAR) mIoU 67.84 #2 of 17 Archive leaderboard report
Semantic Segmentation MCubeS CMNeXt (B2 RGB-A-D-N) mIoU 51.54% #10 of 22 Archive leaderboard report
Semantic Segmentation MCubeS CMNeXt (B2 RGB-A-D) mIoU 49.48% #17 of 22 Archive leaderboard report
Semantic Segmentation MCubeS CMNeXt (B2 RGB-A) mIoU 48.42% #18 of 22 Archive leaderboard report
Semantic Segmentation MCubeS (P) CMNeXt (B2 RGB-A-D) mIoU 49.48 #7 of 8 Archive leaderboard report
Semantic Segmentation MCubeS (P) CMNeXt (B2 RGB-A) mIoU 48.42 #8 of 8 Archive leaderboard report
Semantic Segmentation NYU Depth v2 CMNeXt (B4) Mean IoU 56.9% #18 of 121 Archive leaderboard report
Semantic Segmentation Porto CMNeXt IoU 73.12 #1 of 6 Archive leaderboard report
Semantic Segmentation TLCGIS CMNeXt IoU 82.26 #4 of 6 Archive leaderboard report
Semantic Segmentation UrbanLF CMNeXt (RGB-LF80) mIoU (Real) 83.11 #1 of 14 Archive leaderboard report
Semantic Segmentation UrbanLF CMNeXt (RGB-LF80) mIoU (Syn) 81.02 #1 of 14 Archive leaderboard report
Semantic Segmentation UrbanLF CMNeXt (RGB-LF33) mIoU (Real) 82.62 #2 of 14 Archive leaderboard report
Semantic Segmentation UrbanLF CMNeXt (RGB-LF33) mIoU (Syn) 80.98 #2 of 14 Archive leaderboard report
Semantic Segmentation UrbanLF CMNeXt (RGB-LF8) mIoU (Real) 83.22 #3 of 14 Archive leaderboard report
Semantic Segmentation UrbanLF CMNeXt (RGB-LF8) mIoU (Syn) 80.74 #3 of 14 Archive leaderboard report
Thermal Image Segmentation MFN Dataset CMNeXt (B4) mIOU 59.9 #6 of 55 Archive leaderboard report
Thermal Image Segmentation Noisy RS RGB-T Dataset CMNeXt (B4) mIoU 60.3 #1 of 6 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.

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