Papers › SMD-Nets: Stereo Mixture Density Networks

SMD-Nets: Stereo Mixture Density Networks

8 Apr 2021CVPR 2021 1arXiv:2104.03866archive 2025-07-28

Fabio Tosi, Yiyi Liao, Carolin Schmitt, Andreas Geiger

Despite stereo matching accuracy has greatly improved by deep learning in the last few years, recovering sharp boundaries and high-resolution outputs efficiently remains challenging. In this paper, we propose Stereo Mixture Density Networks (SMD-Nets), a simple yet effective learning framework compatible with a wide class of 2D and 3D architectures which ameliorates both issues. Specifically, we exploit bimodal mixture densities as output representation and show that this allows for sharp and precise disparity estimates near discontinuities while explicitly modeling the aleatoric uncertainty inherent in the observations. Moreover, we formulate disparity estimation as a continuous problem in the image domain, allowing our model to query disparities at arbitrary spatial precision. We carry out comprehensive experiments on a new high-resolution and highly realistic synthetic stereo dataset, consisting of stereo pairs at 8Mpx resolution, as well as on real-world stereo datasets. Our experiments demonstrate increased depth accuracy near object boundaries and prediction of ultra high-resolution disparity maps on standard GPUs. We demonstrate the flexibility of our technique by improving the performance of a variety of stereo backbones.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

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 28 code samples harvested from 2 repositories linked to this paper: 15 ran (of which 3 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 5 where Syntology's instrument failed) · 13 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 5 ran (3 with no instrument failure; 2 where Syntology's instrument failed); community (archive-listed): 14 samples from 1 repository, 10 ran (7 with no instrument failure; 3 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

fabiotosi92/SMD-Nets officialmentioned in paperpytorchSyntology: 14 samples, 5 ran (2 instrument-failed) · 9 unverified report
Mirmix/ddlmvs mentioned on GitHubpytorchSyntology: 14 samples, 10 ran (3 instrument-failed) · 4 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

28 samples harvested: 15 ran (of which 3 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 5 where Syntology's instrument failed) · 13 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · our draft was wrong
1ran; the contract check could not run
10ran
13unverified

Licence: 0 of the 28 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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”. For 3 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

fabiotosi92/SMD-Nets official repository 14 samples, 5 ran (2 instrument-failed) · 9 unverified
Regressor in 8 papers lib/model/SMDHead.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · c2896a15e4b9aa7d · report fixture: returned Regressor · level recorded 2026-09-22
ResNetMultiImageInput lib/model/SMDHead.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 0e653694acc4e981 · report fixture: returned ResNetMultiImageInput · level recorded 2026-09-22
Sine in 13 papers lib/model/SMDHead.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · a16de707e5586b16 · report fixture: returned Sine · level recorded 2026-09-22
local_contrast_normalization lib/model/SMDHead.py ran · our draft was wrong MIT (permissive) · eb4a51cd37197a6e · report fixture: returned tuple · level recorded 2026-09-22
resnet_multiimage_input in 13 papers lib/model/SMDHead.py ran · our draft was wrong MIT (permissive) · 98f5a5eb6fec0a91 · report fixture: returned ResNetMultiImageInput · level recorded 2026-09-22
DepthDecoder in 6 papers lib/model/SMDHead.py unverified MIT (permissive) · 7ccf369a6cafb07a · report
HSMNet lib/model/SMDHead.py unverified MIT (permissive) · 338810fc5e2043a8 · report
PSMNet lib/model/SMDHead.py unverified MIT (permissive) · 15bbcc83e8f87378 · report
ResnetEncoder lib/model/SMDHead.py unverified MIT (permissive) · 4af63eb5010aaaf7 · report
SDFilter lib/model/SMDHead.py unverified MIT (permissive) · 12cf0361c050441b · report
SMDHead lib/model/SMDHead.py unverified MIT (permissive) · c8a3ac84df7ddea3 · report
gaussian_blur in 17 papers lib/model/SMDHead.py unverified MIT (permissive) · e2fb247c688205e5 · report
get_backbone in 28 papers lib/model/SMDHead.py unverified MIT (permissive) · 23b44af6dc3ed7f3 · report
hourglass in 7 papers lib/model/SMDHead.py unverified MIT (permissive) · 47c7f7aa6bec2802 · report
Mirmix/ddlmvs community (archive-listed) 14 samples, 10 ran (3 instrument-failed) · 4 unverified
ConvBnReLU3D models/net.py ran · metamorphic tier: invariant MIT (permissive) · 76a27023b936b897 · report fixture: signature x:Tensor/5/float · returned Tensor · level recorded 2026-09-22
DepthInitialization models/net.py ran MIT (permissive) · 92bc6ab66d10346b · report fixture: signature random_initialization:bool/0/bool, min_depth:Tensor/1/float, max_depth:Tensor/1/float, height:int/0/int, width:int/0/int, depth_interval_scale:float/0/float, device:device/0/?, depth:Tensor/4/float · returned Tensor · level recorded 2026-09-24
FeatureNet in 7 papers models/net.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · ee2cc279bdf6169f · report fixture: signature x:Tensor/4/float · returned dict · level recorded 2026-09-22
FeatureWeightNet models/net.py ran MIT (permissive) · fde70f1c0e22a3e9 · report fixture: signature ref_feature:Tensor/4/float, grid:Tensor/4/float · returned Tensor · level recorded 2026-09-24
PixelwiseNet models/net.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · de28fde5fb92de44 · report fixture: signature x1:Tensor/5/float · returned Tensor · level recorded 2026-09-22
Propagation models/net.py ran · metamorphic tier: invariant MIT (permissive) · 9dc98895fe55dc6b · report fixture: signature batch:int/0/int, height:int/0/int, width:int/0/int, depth_sample:Tensor/4/float, grid:Tensor/4/float, depth_min:Tensor/1/float, depth_max:Tensor/1/float, depth_interval_scale:float/0/float · returned Tensor · level recorded 2026-09-22
SimilarityNet models/net.py ran fingerprinted MIT (permissive) · 52e4096f56488419 · report fixture: signature x1:Tensor/5/float, grid:Tensor/4/float, weight:Tensor/5/float · returned Tensor · level recorded 2026-09-24
depth_regression in 8 papers models/net.py ran · our draft was wrong fingerprinted MIT (permissive) · e8f9986089f1f048 · report fixture: signature p:Tensor/4/float, depth_values:Tensor/2/float · returned Tensor · level recorded 2026-09-22
depth_weight models/net.py ran · our draft was wrong fingerprinted MIT (permissive) · 25252e58e3b331de · report fixture: signature depth_sample:Tensor/4/float, depth_min:Tensor/1/float, depth_max:Tensor/1/float, grid:Tensor/4/float, patchmatch_interval_scale:float/0/float, evaluate_neighbors:int/0/int · returned Tensor · level recorded 2026-09-22
differentiable_warping models/net.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · 00b8d4f5fece378f · report fixture: signature src_fea:Tensor/4/float, src_proj:Tensor/3/float, ref_proj:Tensor/3/float, depth_samples:Tensor/4/float · returned Tensor · level recorded 2026-09-22
DDLMVS models/net.py unverified MIT (permissive) · 89bd4abf9c72da4f · report
Evaluation models/net.py unverified MIT (permissive) · c0aa140f64e3d229 · report
PatchMatch models/net.py unverified MIT (permissive) · 9c7781a0f6f917c4 · report
Refinement models/net.py unverified MIT (permissive) · 117934fc23392abd · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for SMD-Nets (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for SMD-Nets (graph method: SMD-Nets). 14 bricks; individually on a fixture: 5 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 9 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. SMDHead entry · level 0 · unverified · c8a3ac84df7ddea3
  2. Regressor model · level 1 · ran · c2896a15e4b9aa7d
  3. get_backbone helper · level 0 · unverified · 23b44af6dc3ed7f3
  4. Sine helper · level 1 · ran · a16de707e5586b16
  5. DepthDecoder helper · level 0 · unverified · 7ccf369a6cafb07a
  6. HSMNet helper · level 0 · unverified · 338810fc5e2043a8
  7. PSMNet helper · level 0 · unverified · 15bbcc83e8f87378
  8. ResNetMultiImageInput helper · level 1 · ran · 0e653694acc4e981
  9. ResnetEncoder helper · level 0 · unverified · 4af63eb5010aaaf7
  10. SDFilter helper · level 0 · unverified · 12cf0361c050441b
  11. gaussian_blur helper · level 0 · unverified · e2fb247c688205e5
  12. hourglass helper · level 0 · unverified · 47c7f7aa6bec2802
  13. local_contrast_normalization helper · level 1 · ran · our draft was wrong · eb4a51cd37197a6e
  14. resnet_multiimage_input helper · level 1 · ran · our draft was wrong · 98f5a5eb6fec0a91

harvested Assembly for SMD-Nets (graph method: SMD-Nets). 15 bricks; individually on a fixture: 11 ran (of which 1 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 3 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. DDLMVS entry · level 0 · unverified · 89bd4abf9c72da4f
  2. FeatureNet model · level 1 · ran · ee2cc279bdf6169f
  3. PatchMatch model · level 0 · unverified · 9c7781a0f6f917c4
  4. Refinement model · level 0 · unverified · 117934fc23392abd
  5. depth_regression helper · level 1 · ran · our draft was wrong · e8f9986089f1f048
  6. ConvBnReLU helper · level 1 · ran · b5fca0e98890a032
  7. DepthInitialization helper · level 1 · ran · 92bc6ab66d10346b
  8. Evaluation helper · level 0 · unverified · c0aa140f64e3d229
  9. FeatureWeightNet model · level 1 · ran · fde70f1c0e22a3e9
  10. Propagation helper · level 1 · ran · 9dc98895fe55dc6b
  11. depth_weight helper · level 1 · ran · our draft was wrong · 25252e58e3b331de
  12. PixelwiseNet model · level 1 · ran · de28fde5fb92de44
  13. SimilarityNet model · level 1 · ran · 52e4096f56488419
  14. differentiable_warping helper · level 1 · ran; the contract check could not run (RAISES) · 00b8d4f5fece378f
  15. ConvBnReLU3D helper · level 1 · ran · 76a27023b936b897

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2104.03866") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Disparity EstimationStereo Matching

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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