Knowledge Commons of Institute of Automation,CAS
Conditional Generative Neural Decoding with Structured CNN Feature Prediction | |
Du CD(杜长德)1; Du CY(杜长营)2; He HG(何晖光)1 | |
2020 | |
会议名称 | Proceedings of the AAAI Conference on Artificial Intelligence |
会议日期 | 2020-4 |
会议地点 | 美国 |
摘要 | Decoding visual contents from human brain activity is a challenging task with great scientific value. Two main facts that hinder existing methods from producing satisfactory results are 1) typically small paired training data; 2) under-exploitation of the structural information underlying the data. In this paper, we present a novel conditional deep generative neural decoding approach with structured intermediate feature prediction. Specifically, our approach first decodes the brain activity to the multilayer intermediate features of a pretrained convolutional neural network (CNN) with a structured multi-output regression (SMR) model, and then inverts the decoded CNN features to the visual images with an introspective conditional generation (ICG) model. The proposed SMR model can simultaneously leverage the covariance structures underlying the brain activities, the CNN features and the prediction tasks to improve the decoding accuracy and interpretability. Further, our ICG model can 1) leverage abundant unpaired images to augment the training data; 2) self-evaluate the quality of its conditionally generated images; and 3) adversarially improve itself without extra discriminator. Experimental results show that our approach yields state-of-the-art visual reconstructions from brain activities. |
DOI | https://doi.org/10.1609/aaai.v34i03.5647 |
收录类别 | EI |
语种 | 英语 |
是否为代表性论文 | 是 |
七大方向——子方向分类 | 脑机接口 |
国重实验室规划方向分类 | 认知机理与类脑学习 |
是否有论文关联数据集需要存交 | 否 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/51627 |
专题 | 脑图谱与类脑智能实验室_神经计算与脑机交互 |
通讯作者 | He HG(何晖光) |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.Huawei Noah's Ark Lab |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Du CD,Du CY,He HG. Conditional Generative Neural Decoding with Structured CNN Feature Prediction[C],2020. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
AAAI-2020.pdf(1813KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论