DKPMV: Dense Keypoints Fusion from Multi-View RGB Frames for 6D Pose Estimation of Textureless Objects

2025-10-12

AI Summary

Accurate 6D pose estimation of textureless objects remains challenging under depth-deficient conditions. Method: This paper proposes a dense keypoint fusion approach leveraging only multi-view RGB images, structured as a three-stage progressive optimization framework. It introduces an attention-driven keypoint aggregation mechanism and a symmetry-aware training strategy to enhance keypoint localization accuracy and mitigate symmetry-induced ambiguities. Crucially, the method operates without depth input, explicitly modeling multi-view geometric constraints for robust pose estimation. Contribution/Results: Evaluated on the ROBI dataset, our method surpasses all existing state-of-the-art multi-view RGB approaches and outperforms leading RGB-D methods on most metrics, demonstrating superior accuracy and generalization capability for textureless object pose estimation.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

6D pose estimation of textureless objects is valuable for industrial robotic applications, yet remains challenging due to the frequent loss of depth information. Current multi-view methods either rely on depth data or insufficiently exploit multi-view geometric cues, limiting their performance. In this paper, we propose DKPMV, a pipeline that achieves dense keypoint-level fusion using only multi-view RGB images as input. We design a three-stage progressive pose optimization strategy that leverages dense multi-view keypoint geometry information. To enable effective dense keypoint fusion, we enhance the keypoint network with attentional aggregation and symmetry-aware training, improving prediction accuracy and resolving ambiguities on symmetric objects. Extensive experiments on the ROBI dataset demonstrate that DKPMV outperforms state-of-the-art multi-view RGB approaches and even surpasses the RGB-D methods in the majority of cases. The code will be available soon.

Problem

• Estimating 6D poses of textureless objects using multi-view RGB images
• Addressing depth information loss in textureless object pose estimation
• Improving multi-view geometric cues without relying on depth data

Innovation

• Uses only multi-view RGB images for input
• Implements three-stage progressive pose optimization
• Enhances keypoint network with attentional aggregation