AI Paper Atlas

Research map / 3D vision / Human pose estimation

Ada-VNNs: Adaptive Equivariance for Vector Neural Networks

Bing Han, Ruitao Pan, Yumin Chen, Peixin Hong, Weiyuan Liu, Zhibin Zhao, Chenxi Wang, Zhi Zhai

ECCV 2026 · 3D vision › Human pose estimation · approach: Learned camera pose estimation · 0 citations (OpenAlex)

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 the Adaptive Gating Unit and residual pathway to VNNVNN · ShapeNet continuous self-symmetry (avg)10° registration recall: 8.87 -> 70.31Improved
add the Adaptive Gating Unit and residual pathway to VNNVNN · ShapeNet discrete self-symmetry (avg)10° registration recall: 21.29 -> 70.41Improved
add the Adaptive Gating Unit and residual pathway to VNNVNN · ShapeNet asymmetric (avg)10° registration recall: 67.19 -> 86.24Improved
add the Adaptive Gating Unit and residual pathway to VN-TransformerVN-T · ShapeNet continuous self-symmetry (avg)10° registration recall: 51.17 -> 72.09Improved
add the Adaptive Gating Unit and residual pathway to VN-TransformerVN-T · ShapeNet discrete self-symmetry (avg)10° registration recall: 76.10 -> 80.49Improved
add the Adaptive Gating Unit and residual pathway to VN-TransformerVN-T · ShapeNet asymmetric (avg)10° registration recall: 86.10 -> 85.77No clear effect
add the Adaptive Gating Unit and residual pathway to VNNVNN · Isotope prediction (C6H6)angular error (rad): 0.946 -> 0.302Improved
add the adaptive branch to VN-DGCNNVN-DGCNN · ModelNet40 z/SO(3)accuracy: 90.2 -> 86.9Hurt
add the adaptive branch to VN-PointNetVN-PointNet · ModelNet40 z/zaccuracy: 77.5 -> 81.2Improved
add the adaptive branch to VN-DGCNNVN-DGCNN · ModelNet40 invariance errorinvariance error: 0.0023 -> 0.0040Hurt
increase AGLL layers in AGU from 1 to 2 (ablation)Ada-VNN · ShapeNet Lamp10° registration recall: ~30 -> 79.44Improved
increase AGLL layers k in AGU (ablation)Ada-VNN · ShapeNet 26 categoriesequivariance error: roughly one order of magnitude per added layerHurt

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