通用智能本体--人工智能-上海期智研究院

通用智能本体-

卢策吾

课题介绍

我们是针对目前智能设备无法实现在现实中对任务和场景自适应需求的问题，面向开发通用智能本体，追求各种任务间的通用可转移性，实现大部分任务可以在同一个本体上执行，从而大大提高产业和生活的智能程度。目前围绕通用智能本体的开发展开了一系列研究，取得了以下进展：

① 硬件本体的设计、完善元操作流任务定义和元操作知识引擎，整合成一套完整的系统RobotFlow。

② 硬件本体原型生产完成，软件系统已部分开源（包括视觉模块，运动规划模块）。其余模块会在今年后期陆续开源。项目主页：https://wenqiangx.github.io/robotflowproject/ 。”

课题团队成员

熊浩宇

课题科研成果

Ye R, Xu W, Xue Z, et al. H2O: A Benchmark for Visual Human-human Object Handover Analysis[J]. arXiv preprint arXiv:2104.11466, 2021.

Yang L, Zhan X, Li K, et al. CPF: Learning a Contact Potential Field to Model the Hand-object Interaction[J]. arXiv preprint arXiv:2012.00924, 2020.

Shi R, Xue Z, You Y, et al. Skeleton Merger: an Unsupervised Aligned Keypoint Detector[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 43-52.

Li J, Xu C, Chen Z, et al. HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 3383-3393.

Gou M, Fang H S, Zhu Z, et al. RGB Matters: Learning 7-DoF Grasp Poses on Monocular RGBD Images[J]. arXiv preprint arXiv:2103.02184, 2021.

Pang B, Peng G, Li Y, et al. PGT: A Progressive Method for Training Models on Long Videos[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 11379-11389.

Fang H S, Xie Y, Shao D, et al. DIRV: Dense Interaction Region Voting for End-to-End Human-Object Interaction Detection[J]. arXiv preprint arXiv:2010.01005, 2020.

Fang H S, Xie Y, Shao D, et al. DecAug: Augmenting HOI Detection via Decomposition[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2021, 35(2): 1300-1308.

Li Y L, Liu X, Wu X, et al. HOI analysis: Integrating and decomposing human-object interaction[J]. arXiv preprint arXiv:2010.16219, 2020.

Tang J, Xia J, Mu X, et al. Asynchronous interaction aggregation for action detection[C]//European Conference on Computer Vision. Springer, Cham, 2020: 71-87.

Li J, Wang C, Liu W, et al. Hmor: Hierarchical multi-person ordinal relations for monocular multi-person 3d pose estimation[J]. arXiv preprint arXiv:2008.00206, 2020.

Lou Y, You Y, Li C, et al. Human correspondence consensus for 3d object semantic understanding[C]//European Conference on Computer Vision. Springer, Cham, 2020: 496-512.

分享到