An RGB-D Dataset for 6D Pose Estimation of Texture-less Objects (T-LESS)

by No License

An RGB-D Dataset for 6D Pose Estimation of Texture-less Objects (T-LESS)

A dataset acquired with 3 synchronized sensors (Primesense Carmine 1.09, Microsoft Kinect v2, Canon IXUS 950 IS), featuring:* 30 industry-relevant objects: no discriminative color, no texture, often similar in shape, some objects are parts of others.* Training images depicting individual objects against a black background.* Test images originating from 20 test scenes whose complexity varies from simple scenes with several isolated objects to very challenging ones with multiple object instances and a high amount of clutter and occlusion.* Two types of 3D models for each object: a manually created CAD model and a semi-automatically reconstructed one.* Accurate 3D pose for all object instances and images.Reference:T. Hodaò, P. Haluza, Š. Obdržálek, J. Matas, M. Lourakis, X. Zabulis,T-LESS: An RGB-D Dataset for 6D Pose Estimation of Texture-less Objects,IEEE Winter Conference on Applications of Computer Vision (WACV), pp. 880-888, 2017, Santa Rosa, USA

Dataset Attributes

Label SVG
CategoriesRgbd, 3D, Pose, Texture-Less, Object, Estimation