报告题目：ICASSP2023-Decoupled Visual Causality for Robust Detection
报告华体会网页版登陆入口：2023年6月16日 19:00 - 20:00
报告地点：中南大学铁道学院电子楼414
腾讯会议：914-8957-5796
报告摘要： This work is accepted as an oral paper in the ICASSP2023 (CCF B). The abstract of the accepted paper is as follows. The existing empirical risk minimization algorithms learn the association between inputs and labels, and face substantial difficulties when apply to different distributions because of various confounders. Causal intervention becomes a solid solution to this issue by analyzing the visual causality, instead, those approaches fail at disentangling the confounders and mediators within the causality, and bring negative effects to the prediction. In this paper, we propose a disentangled visual causal model to eliminate the effects of confounders while reserving the corresponding mediators. Specifically, confounders are considered as different objects on the image, while mediators are formulated as some critical components of the targets that contribute to a distinctive identification. Extensive experiments on COCO datasets have demonstrated the superiority of our model over other state-of-the-art baselines.
讲座者简介： Ping Jiang received the master's degree in computer engineering from the University of Western Ontario, Canada, in 2018. He is currently pursuing a Ph.D. degree with the Electrical and Communication Engineering Department, Central South University, China. His research interests include machine learning, computer vision, and perception in vehicular networks.
