AI Paper Atlas

Research map / 3D vision / Human pose estimation

F2G-Pose: Geometry-Aware Foundation Feature Lifting for Direct RGB-D Category-Level Object Pose Estimation

Jinyu Zhang, Haitao Lin, Jiashu Hou, Xiangyang Xue, Yanwei Fu

NeurIPS 2026 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes F2G-Pose, a geometry-aware method that lifts foundation-model features into 3D for direct RGB-D category-level object pose estimation.

Problem

Category-level RGB-D object pose estimation (recovering 3D rotation and translation without instance CAD models or reference views) is ambiguous due to intra-class variation, symmetry, and partial observations; deterministic methods suffer error propagation from intermediate structures, while other

Approach

Lifts features from a foundation model into geometry-aware representations for direct pose regression from RGB-D input, avoiding error-prone intermediate structures; only the abstract is available, so specifics are unknown.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add lifted RADIO RGB foundation features (ablation)F2G-Pose · ROPEAUC IoU50: 16.2 -> 27.8Improved
Add MoE Transformer instead of dense FFN (ablation)F2G-Pose · ROPEAUC IoU50: 24.0 -> 27.8Improved
Add camera-frame shape supervision (ablation)F2G-Pose · ROPEAUC IoU50: 23.2 -> 27.8Improved
Use DGCNN tokens instead of global PointNet pooling (ablation)F2G-Pose · ROPEAUC IoU50: 17.8 -> 21.6Improved
Add Transformer context to DGCNN tokens (ablation)F2G-Pose · ROPEAUC IoU50: 21.6 -> 24.0Improved
Swap RADIO backbone for DINOv2 (ablation)F2G-Pose · ROPEAUC IoU50: 27.8 -> 27.0No clear effect
Swap RADIO backbone for CLIP (ablation)F2G-Pose · ROPEAUC IoU50: 27.8 -> 26.3No clear effect
Use top-2 MoE instead of top-1 (ablation)F2G-Pose · ROPEAUC IoU50: 25.6 -> 27.8Improved
Use top-4 MoE instead of top-2 (ablation)F2G-Pose · ROPEAUC IoU50: 27.8 -> 28.0No clear effect
Add load-balancing loss to top-2 MoE (ablation)F2G-Pose · ROPEAUC IoU50: 26.0 -> 27.8Improved
Replace GenPose++ sampling with direct single-pass predictionF2G-Pose · SOPEAUC IoU50: 31.9 -> 43.7Improved
Replace GenPose++ sampling with direct single-pass prediction, speedF2G-Pose · ROPEFPS: 4 -> 28Improved

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