Disentangling Before Composing: Learning Invariant Disentangled Features for Compositional Zero-Shot Learning

不变(物理) 人工智能 计算机科学 零(语言学) 模式识别(心理学) 数学 数学物理 哲学 语言学
作者
Tian Zhang,Kongming Liang,Ruoyi Du,Wei Chen,Zhanyu Ma
出处
期刊:IEEE Transactions on Pattern Analysis and Machine Intelligence [IEEE Computer Society]
卷期号:: 1-16
标识
DOI:10.1109/tpami.2024.3487222
摘要

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 its corresponding 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 compositions. 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 attributes attached to any object reliably, and vice versa. 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. To further facilitate and encourage the disentanglement of attributes and objects, we propose an "encoding-reshuffling-decoding" process to help the model avoid spurious correlations by randomly regrouping the disentangled features into synthetic features. Ultimately, our method improves generalization by learning to disentangle features that represent two independent factors of attributes and objects. Experiments demonstrate that the proposed method achieves state-of-the-art or competitive performance in both closed-world and open-world scenarios. Codes are available at https://github.com/PRIS-CV/Disentangling-before-Composing.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
gjm完成签到,获得积分10
1秒前
SciGPT应助zj采纳,获得10
2秒前
Ava应助阿湫采纳,获得10
2秒前
meng123完成签到,获得积分20
3秒前
x5kyi完成签到,获得积分10
4秒前
爆米花应助肖遥采纳,获得10
5秒前
Xx完成签到,获得积分10
5秒前
5秒前
8秒前
烟里戏完成签到 ,获得积分10
10秒前
shuangfeng1853完成签到 ,获得积分10
10秒前
林子青发布了新的文献求助10
10秒前
11秒前
aa完成签到,获得积分10
11秒前
CXC完成签到,获得积分10
11秒前
13秒前
Zzz发布了新的文献求助10
13秒前
上官若男应助袁凯文采纳,获得10
14秒前
14秒前
褚晣完成签到,获得积分10
14秒前
ATTENTION完成签到,获得积分10
15秒前
15秒前
周欣玙完成签到,获得积分10
15秒前
15秒前
学不懂数学应助as采纳,获得30
15秒前
传奇3应助YiWei采纳,获得10
16秒前
阿湫发布了新的文献求助10
16秒前
why完成签到,获得积分10
16秒前
16秒前
慕青应助Zard采纳,获得10
19秒前
肖遥发布了新的文献求助10
20秒前
20秒前
天气一级棒完成签到,获得积分10
21秒前
TAA66完成签到,获得积分10
21秒前
小凯同学完成签到,获得积分10
21秒前
21秒前
星辰完成签到,获得积分10
24秒前
gomm完成签到,获得积分10
24秒前
星辰大海应助票子采纳,获得10
25秒前
高分求助中
【提示信息,请勿应助】关于scihub 10000
Les Mantodea de Guyane: Insecta, Polyneoptera [The Mantids of French Guiana] 3000
徐淮辽南地区新元古代叠层石及生物地层 3000
The Mother of All Tableaux: Order, Equivalence, and Geometry in the Large-scale Structure of Optimality Theory 3000
Global Eyelash Assessment scale (GEA) 1000
Picture Books with Same-sex Parented Families: Unintentional Censorship 550
Research on Disturbance Rejection Control Algorithm for Aerial Operation Robots 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 4038524
求助须知:如何正确求助?哪些是违规求助? 3576221
关于积分的说明 11374737
捐赠科研通 3305912
什么是DOI,文献DOI怎么找? 1819354
邀请新用户注册赠送积分活动 892688
科研通“疑难数据库(出版商)”最低求助积分说明 815048