We propose FoundPose, a model-based method for 6D pose estimation of unseen objects from a single RGB image. The method can quickly onboard new objects using their 3D models without requiring any object- or task-specific training. In contrast, existing methods typically pre-train on large-...
Abstract Current 6D object pose estimation methods usually require a 3D model for each object. These methods also require additional training in order to incorporate new objects. As a result, they are difficult to scale to a large number of objects and cannot be directly applied to unseen object...
We propose FoundPose, a model-based method for 6D pose estimation of unseen objects from a single RGB image. The method can quickly onboard new objects using their 3D models without requiring any object- or task-specific training. In contrast, existing methods typically pre-train on large-scal...
Implementation of "FoundPose Unseen Object Pose Estimation with Foundation Features", ECCV 2024 - facebookresearch/foundpose
@inproceedings{park2019latentfusion,title={LatentFusion: End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimation},author={Park, Keunhong and Mousavian, Arsalan and Xiang, Yu and Fox, Dieter},booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern ...
原 论文阅读--CVPR2018--video object segmentation--2 , motion sensor (GPS/IMU) and a3Dsemantic map together. The first step is cameraposeestimation...DeLS-3D: Deep Localization and Segmentationwitha3DSemantic Map 略读,motivation This research ...
Kernelized few-shot object detection with efficient integral aggregation. In IEEE Conference on Computer Vision and Pattern Recognition, 2022. 2 [56] Teng Zhang, Liangchen Liu, Kun Zhao, Arnold Wiliem, Gra- ham Hemson, and Brian Lovell. Omni-supervised joint de-...
Currently, most 3D perception tasksare passive, such as object detection [1]–[3], object poseestimation [4, 5], object reconstruction [6, 7], etc. Thesemethods either rely on known object models or large amountsof annotated data for training, which limits their applicabilityin the real ...
We evaluate the performance of our method for unseen object pose estimation on MOPED as well as the ModelNet and LINEMOD datasets. Our method performs competitively to supervised methods that are trained on those objects. Code and data will be available at https://keunhong.com/publications/...
Six degrees of freedom pose estimation technology constitutes the cornerstone for precise robotic control and similar tasks. Addressing the limitations of current 6-DoF pose estimation methods in handling object occlusions and unknown objects, we have developed a novel two-sta...