The self-attention networks and Transformer have dominated machine translation and natural language processing fields,and shown great potential in image vision tasks such as image classification and object detection.I...The self-attention networks and Transformer have dominated machine translation and natural language processing fields,and shown great potential in image vision tasks such as image classification and object detection.Inspired by the great progress of Transformer,we propose a novel general and robust voxel feature encoder for 3D object detection based on the traditional Transformer.We first investigate the permutation invariance of sequence data of the self-attention and apply it to point cloud processing.Then we construct a voxel feature layer based on the self-attention to adaptively learn local and robust context of a voxel according to the spatial relationship and context information exchanging between all points within the voxel.Lastly,we construct a general voxel feature learning framework with the voxel feature layer as the core for 3D object detection.The voxel feature with Transformer(VFT)can be plugged into any other voxel-based 3D object detection framework easily,and serves as the backbone for voxel feature extractor.Experiments results on the KITTI dataset demonstrate that our method achieves the state-of-the-art performance on 3D object detection.展开更多
With this article we aim to propose an analytical alternative to what we could call the substitution and condemnation of the spectator as an element of the new technological apparatuses. Actually, what we see in many ...With this article we aim to propose an analytical alternative to what we could call the substitution and condemnation of the spectator as an element of the new technological apparatuses. Actually, what we see in many of the current reflections on new media is that while the spectator is being accused of ultimate passivity, he is also being mobilized to emancipated action as an empowered user. Through the critique of an online application available on Portuguese Radio Television （RTP）＇s website, the state-owned television network, called 0 Meu Telejornal1, we will try to demonstrate that we actually need to develop a spectator theory to critically understand the political position of the citizen within our changing contemporary media environment.展开更多
基金National Natural Science Foundation of China(No.61806006)Innovation Program for Graduate of Jiangsu Province(No.KYLX160-781)University Superior Discipline Construction Project of Jiangsu Province。
文摘The self-attention networks and Transformer have dominated machine translation and natural language processing fields,and shown great potential in image vision tasks such as image classification and object detection.Inspired by the great progress of Transformer,we propose a novel general and robust voxel feature encoder for 3D object detection based on the traditional Transformer.We first investigate the permutation invariance of sequence data of the self-attention and apply it to point cloud processing.Then we construct a voxel feature layer based on the self-attention to adaptively learn local and robust context of a voxel according to the spatial relationship and context information exchanging between all points within the voxel.Lastly,we construct a general voxel feature learning framework with the voxel feature layer as the core for 3D object detection.The voxel feature with Transformer(VFT)can be plugged into any other voxel-based 3D object detection framework easily,and serves as the backbone for voxel feature extractor.Experiments results on the KITTI dataset demonstrate that our method achieves the state-of-the-art performance on 3D object detection.
文摘With this article we aim to propose an analytical alternative to what we could call the substitution and condemnation of the spectator as an element of the new technological apparatuses. Actually, what we see in many of the current reflections on new media is that while the spectator is being accused of ultimate passivity, he is also being mobilized to emancipated action as an empowered user. Through the critique of an online application available on Portuguese Radio Television （RTP）＇s website, the state-owned television network, called 0 Meu Telejornal1, we will try to demonstrate that we actually need to develop a spectator theory to critically understand the political position of the citizen within our changing contemporary media environment.
