AI Paper Atlas

Research map / 3D vision / Depth and stereo

Deep Two-View Structure-From-Motion Revisited

Jianyuan Wang, Yiran Zhong, Yuchao Dai, Stan Birchfield, Kaihao Zhang, Nikolai Smolyanskiy, Hongdong Li

CVPR 2021 · 3D vision › Depth and stereo · approach: Feed-forward 3D reconstruction · 54 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use the classic correspondence-to-pose pipeline with scale-invariant depth matchingDeep two-view SfM · KITTI Depth EigenRMSE: 2.946 -> 2.273Improved
use the proposed two-view SfM pipeline rather than DeepV2DKITTI Depth Eigen SfM · KITTI Depth Eigen SfMD1-all: 9.1% vs 3.1%Improved
use the proposed deep correspondence and geometric pose pipelineTwo-view SfM · KITTI VO Seq. 09t_err (%): 3.49 -> 1.70Improved
use the proposed deep correspondence and geometric pose pipelineTwo-view SfM · KITTI VO Seq. 10t_err (%): 5.81 -> 1.49Improved
add rigid-flow supervision to fine-tune optical flow (ablation)Our optical flow estimator · MVSTran: 10.984 -> 3.878Improved
add rigid-flow supervision to fine-tune optical flow (ablation)Our optical flow estimator · Scenes11Tran: 6.617 -> 2.041Improved
add rigid-flow supervision to fine-tune optical flow (ablation)Our optical flow estimator · SUN3DTran: 12.905 -> 10.757Improved
mask optical-flow matches using SIFT feature locations before five-point RANSAC (ablation)Five-point pose estimation · MVSTran: 37.292 -> 3.878Improved
use scale-invariant plane-sweep matching (ablation)Depth estimation pipeline · KITTI Depth EigenRMSE: 3.120 -> 2.273Improved
replace scale-invariant matching with a scale-invariant loss (ablation)Depth estimation pipeline · KITTI Depth EigenRMSE: 2.273 -> 2.981Hurt
train with ground-truth pose despite the scale mismatch (ablation)Depth estimation pipeline · KITTI Depth EigenRMSE: 3.120 -> 3.421Hurt
apply scale-invariant matching to DeepV2D (ablation)DeepV2D · KITTI Depth EigenAbs Rel: 0.064 -> 0.063No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy