VALSE 首页 活动通知 查看内容

VALSE 论文速览 第99期:学习不变性视觉表示的零样本组合学习 ...

2022-9-29 10:40| 发布者: 程一-计算所| 查看: 770| 评论: 0

摘要: 为了使得视觉与学习领域相关从业者快速及时地了解领域的最新发展动态和前沿技术进展,VALSE最新推出了《论文速览》栏目,将在每周发布一至两篇顶会顶刊论文的录制视频,对单个前沿工作进行细致讲解。本期VALSE论文速 ...


论文题目:Learning Invariant Visual Representations for Compositional Zero-Shot Learning

作者列表:张甜 (北京邮电大学)、梁孔明 (北京邮电大学)、杜若一 (北京邮电大学)、孙显 (中国科学院航空航天信息研究所)、马占宇 (北京邮电大学)、郭军 (北京邮电大学)



Compositional Zero-Shot Learning (CZSL)aims to recognize novel compositions using knowledge learned from seen attribute-object compositions in the training set. Previous works mainly project an image and a composition into a common embedding space to measure their compatibility score. However, both attributes and objects share the visual representations learned above, leading the model to exploit spurious correlations and bias towards seen pairs. Instead, we reconsider CZSL as an out-of-distribution generalization problem. If an object is treated as a domain, we can learn object-invariant features to recognize the attributes attached to any object reliably. Similarly, attribute-invariant features can also be learned when recognizing the objects with attributes as domains. Specifically, we propose an invariant feature learning framework to align different domains at the representation and gradient levels to capture the intrinsic characteristics associated with the tasks. Experiments on three CZSL benchmarks demonstrate that the proposed method significantly outperforms the previous state-of-the-art.


[1] Tian Zhang, Kongming Liang, Ruoyi Du, Xian Sun, Zhanyu Ma, and Jun Guo. Learning Invariant Visual Representations for Compositional Zero-Shot Learning. ECCV 2022.








月度轮值AC:王立君 (大连理工大学)、眭亚楠 (清华大学)

季度责任AC:张姗姗 (南京理工大学)

小黑屋|手机版|Archiver|Vision And Learning SEminar

GMT+8, 2023-12-11 16:20 , Processed in 0.011468 second(s), 14 queries .

Powered by Discuz! X3.4

Copyright © 2001-2020, Tencent Cloud.