Meta tags:

Headings (most frequently used words):

dataset, level, synthetic, articulation, view, point, random, real, category, articulated, object, pose, estimation, cvpr, 2020, oral, video, results, paper, team, bibtex, acknowledgements, contact, continuous, fixed, simu, to, instance,

Text of the page (most frequently used words):
and (13), the (9), for (7), #category (7), pose (6), level (6), object (6), research (5), from (5), space (5), canonical (5), part (5), articulated (4), estimation (4), virginia (3), tech (3), wang (3), xiaolong (3), this (3), arxiv (3), dataset (3), articulation (3), joint (3), stanford (2), university (2), google (2), contact (2), edu (2), guibas (2), leonidas (2), abbott (2), lynn (2), song (2), shuran (2), 1912 (2), 11913 (2), 2020 (2), paper (2), real (2), synthetic (2), random (2), view (2), point (2), objects (2), single (2), depth (2), that (2), ancsh (2), representation (2), normalizes (2), parameters (2), scale (2), joints (2), meet, cardboard, robot, danbo, columbia, you, have, any, questions, please, feel, free, lxiaol9_at_vt, hewang_at_stanford, was, supported, grant, toyota, center, used, resources, provided, advanced, computing, within, division, information, technology, thank, vision, learning, lab, help, visualization, tools, are, also, grateful, financial, hardware, support, acknowledgements, article, li2019articulated, title, author, journal, preprint, year, 2019, bibtex, stands, equal, contribution, team, latest, version, march, here, simu, instance, continuous, fixed, results, video, release, april, code, pretrained, model, data, project, addresses, task, image, present, novel, approach, correctly, accommodates, instances, previously, unseen, during, training, introduce, aware, normalizedcoordinate, hierarchy, different, given, key, achieve, intra, generalization, constructs, objectspace, well, set, spaces, thecanonical, orientation, scales, articulations, states, while, each, further, its, develop, deep, network, based, pointnet, predicts, pointcloud, including, segmentation, normalized, coordinates, leveraging, canonicalized, demonstrate, improved, performance, estimations, using, induced, kinematic, constraints, high, accuracy, parameter, camera, cvpr, oral,

Text of the page (random words):
category level category level articulated object pose estimation cvpr 2020 oral this project addresses the task of category level pose estimation for articulated objects from a single depth image we present a novel category level approach that correctly accommodates object instances previously unseen during training we introduce articulation aware normalizedcoordinate space hierarchy ancsh a canonical representation for different articulated objects in a given category as the key to achieve intra category generalization the representation constructs a canonical objectspace as well as a set of canonical part spaces thecanonical object space normalizes the object orientation scales and articulations e g joint parameters and states while each canonical part space further normalizes its part pose and scale we develop a deep network based on pointnet that predicts ancsh from a single depth pointcloud including part segmentation normalized coordinates and joint parameters in the canonical object space by leveraging the canonicalized joints we demonstrate 1 improved performance in part pose and scale estimations using the induced kinematic constraints from joints 2 high accuracy for joint parameter estimation in camera space paper code pretrained model data release in april video results synthetic dataset continuous articulation fixed view point synthetic dataset random articulation random view point real dataset simu to real instance level paper latest version march 31 2020 arxiv 1912 11913 in cs cv or here team xiaolong li 1 he wang 2 li yi 3 leonidas j guibas 2 a lynn abbott 1 shuran song 4 1 virginia tech 2 stanford university 3 google research 4 columbia university stands for equal contribution bibtex article li2019articulated pose title category level articulated object pose estimation author li xiaolong and wang he and yi li and guibas leonidas and abbott a lynn and song shuran journal arxiv preprint arxiv 1912 11913 year 2019 acknowledgements this research was supported by a grant from toyota stanford center for ai research this research used resources provided by advanced research computing within the division of information technology at virginia tech we thank vision and learning lab at virginia tech for help on visualization tools we are also grateful for financial and hardware support from google contact if you have any questions please feel free to contact xiaolong li at lxiaol9_at_vt edu and he wang at hewang_at_stanford edu meet danbo the cardboard robot