DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusion

DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusion
Chen Wang, Danfei Xu, Yuke Zhu, Roberto Martín-Martín, Cewu Lu, Li Fei-Fei, Silvio Savarese
Website: sites.google.c...
Abstract:
A key technical challenge in performing 6D object pose estimation from RGB-D image is to fully leverage the two complementary data sources. Prior works either extract information from the RGB image and depth separately or use costly post-processing steps, limiting their performances in highly cluttered scenes and real-time applications. In this work, we present DenseFusion, a generic framework for estimating 6D pose of a set of known objects from RGB-D images. DenseFusion is a heterogeneous architecture that processes the two data sources individually and uses a novel dense fusion network to extract pixel-wise dense feature embedding, from which the pose is estimated. Furthermore, we integrate an end-to-end iterative pose refinement procedure that further improves the pose estimation while achieving near real-time inference. Our experiments show that our method outperforms state-of-the-art approaches in two datasets, YCB-Video and LineMOD. We also deploy our proposed method to a real robot to grasp and manipulate objects based on the estimated pose.

[@renanmoreira7506]

looks amazing but only it because still not open to public not even the examples are working ;/ tryed to use but only crash when u use a camera

[@junione1997]

is it also possible to estimate the length and width of an object? how so?

[@kayumiy]

What does " known objects" mean?

[@serbianhero1389]

Full 3D model of an object is available before the network is trained