AI Paper Atlas

Research map / 3D vision / Human pose estimation

DeepSFM: Structure From Motion Via Deep Bundle Adjustment

Xingkui Wei, Yinda Zhang, Zhuwen Li, Yanwei Fu, Xiangyang Xue

ECCV 2020 · 3D vision › Human pose estimation · approach: Category-level object pose estimation

Read the paperCheck 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
add geometric (depth) consistency channels to cost volumes (ablation)DeepSFM · MVSabs_rel: 0.0813 -> 0.0698Improved
add geometric consistency channels (ablation)DeepSFM · MVSsq_rel: 0.0971 -> 0.0523Improved
use nearest neighbor instead of bilinear sampling for depth warping (ablation)DeepSFM · DeMoN MVSL1-inv: 0.023 -> 0.021Improved
use nearest neighbor instead of bilinear sampling for depth warping (ablation)DeepSFM · DeMoN MVSTrans error: 9.910 -> 9.881Improved
replace direct visual odometry pose update with P-CV (ablation)DeepSFM · DeMoN iterationsdepth and pose errorImproved
iterate refinement from DeMoN initialization up to iteration 4 (ablation)DeepSFM · DeMoN iterations (test)abs relative: 0.254 -> 0.126Improved
run more than 4 test iterations (up to 20) (ablation)DeepSFM · DeMoN iterations (test)abs relative: 0.126 -> 0.120No clear effect
add scale invariant gradient loss (ablation)DeepSFM · MVSabs_rel: 0.0698 -> 0.0712No clear effect
add small Gaussian noise to initialization during fine-tuning (ablation)DeepSFM · MVSdepth and pose accuracyHurt
use translation bin size 0.10 instead of 0.05 (ablation)DeepSFM · MVSRot error: 3.308 -> 2.825Improved
use rotation bin 0.03 instead of 0.07 (ablation)DeepSFM · MVSRot error: 3.024 -> 2.825Improved
use DeepSFM instead of COLMAPCOLMAP · ETH3Dsq_rel: 36.71 -> 0.278Improved

Later papers that compared with this method

2 later accepted papers report DeepSfM as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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