Abstract: ?
Eliminating the redundancy in convolution kernels has been ?attracting increasing interests for designing efficient convolutional ?neural network architectures with three goals: small model, fast ?computation, and high accuracy. Existing solutions include ?low-precision kernels, structured sparse kernels, low-rank kernels, ?and the product of low-rank kernels. In this talk, I will introduce a ?novel framework: Interleaved Group Convolution (IGC), which uses the ?product of structured sparse kernels to compose a dense convolution ?kernel. It is a drop-in replacement of normal convolution and can be ?applied to any networks that depend on convolution. I present the ?complementary condition and the balance condition to guide the design, ?obtaining a balance between three aspects: model size, computation ?complexity and classification performance. I will show empirical and ?theoretic justification of the advantage of the proposed approach over ?Xception and MobileNet. In addition, IGC raises a rarely-studied ?matrix decomposition problem: sparse matrix factorization (SMF). I ?expect more research efforts in SMF from the researchers in the area ?of matrix analysis.
?About speaker: ?
Jingdong Wang is a Senior Researcher with the Visual Computing ?Group, Microsoft Research, Beijing, China. His areas of current ?interest include efficient CNN architecture design, person ?re-identification, human pose estimation, semantic segmentation, ?large-scale indexing, and salient object detection. He has authored ?one book and 100+ papers in top conferences and prestigious ?international journals in computer vision, multimedia, and machine ?learning. His paper was selected into the Best Paper Finalist at the ?ACM MM 2015. He has shipped a dozen of technologies to Microsoft ?products, including Bing search, Cognitive service, and XiaoIce ?Chatbot. Dr. Wang is an Associate Editor of IEEE TPAMI, IEEE TCSVT and ?IEEE TMM. He was an Area Chair or a Senior Program Committee Member of ?top conferences, such as CVPR, ICCV, ECCV, AAAI, IJCAI, and ACM ?Multimedia. He is an ACM Distinguished Member and a Fellow of the IAPR.
?报告人简介: ?
王井东，微软亚洲研究中心视觉计算组高级研究员。王井东研究员的主要研究领域包括CNN架构设计、再度识别、人体姿态估计、语义分割、大规模索引和显著目标检测等，出版著作1部并在计算机视觉、多媒体和机器学习领域相关顶级会议和著名国际期刊发表论文100多篇，曾入选2015年ACM ?MM决赛最佳论文。在技术方面，为微软产品提供了多项技术，包括必应搜索、认知服务和小冰聊天机器人。同时，王井东研究员担任国际核心期刊IEEE ?TPAMI、IEEE TCSVT和IEEE TMM副主编，并曾任CVPR、ICCV、ECCV、AAAI、IJCAI、ACM Multimedia等顶级国际会议的区域主席或资深项目委员会成员。目前是ACM的杰出成员，也是IAPR的成员之一。