Timezone: »

 
Poster
Transductive Zero-Shot Learning with Visual Structure Constraint
Ziyu Wan · Dongdong Chen · Yan Li · Xingguang Yan · Junge Zhang · Yizhou Yu · Jing Liao

Wed Dec 11 05:00 PM -- 07:00 PM (PST) @ East Exhibition Hall B + C #23

To recognize objects of the unseen classes, most existing Zero-Shot Learning (ZSL) methods first learn a compatible projection function between the common semantic space and the visual space based on the data of source seen classes, then directly apply it to the target unseen classes. However, in real scenarios, the data distribution between the source and target domain might not match well, thus causing the well-known domain shift problem. Based on the observation that visual features of test instances can be separated into different clusters, we propose a new visual structure constraint on class centers for transductive ZSL, to improve the generality of the projection function (\ie alleviate the above domain shift problem). Specifically, three different strategies (symmetric Chamfer-distance,Bipartite matching distance, and Wasserstein distance) are adopted to align the projected unseen semantic centers and visual cluster centers of test instances. We also propose a new training strategy to handle the real cases where many unrelated images exist in the test dataset, which is not considered in previous methods. Experiments on many widely used datasets demonstrate that the proposed visual structure constraint can bring substantial performance gain consistently and achieve state-of-the-art results.

Author Information

Ziyu Wan (City University of Hong Kong)
Dongdong Chen (university of science and technology of china)
Yan Li (Institute of Automation, Chinese Academy of Sciences)
Xingguang Yan (Shenzhen University)
Junge Zhang (CASIA)
Yizhou Yu (Deepwise AI Lab)
Jing Liao (City University of Hong Kong)