arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.10018cs.SDcs.MMeess.AS
收藏

基于模态翻译网络的图像音频描述

Audio Description from Image by Modal Translation Network

Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences(中国科学院西安光学精密机械研究所);University of Chinese Academy of Sciences(中国科学院大学);Northwestern Polytechnical University(西北工业大学)

Hailong Ning, Xiangtao Zheng, Yuan Yuan, Xiaoqiang Lu

摘要:

音频是视障者获取信息的主要形式。在现实中,各种视觉数据总是存在的,但音频数据在许多情况下并不存在。为了帮助视觉障碍者更好地感知周围的信息,本文提出了一种图像到音频描述(I2 AD)任务来从图像生成音频描述。为了完成这一全新的任务,提出了从视觉到听觉的模态翻译网络(MT-Net)。所提出的MT-Net包括三个渐进子网络:1)特征学习,2)跨模态映射和3)音频生成。首先,特征学习子网络旨在从图像和音频中学习语义特征,包括图像特征学习和音频特征学习。其次,跨模态映射子网络将图像特征转换为与音频特征具有相同语义概念的跨模态表示。通过这种方式,有效地挖掘了模态间数据的相关性,以缓解图像和音频之间的异构性差距。最后,设计了音频生成子网络,用于从交叉模态表示中生成音频波形。根据采样频率对生成的音频波形进行插值,得到相应的音频文件。作为探索I2 AD任务的第一次尝试,构建了三个具有大量手动音频描述的大规模数据集。实验结果验证了直接从图像中生成清晰语音的可行性和方法的有效性。

英文摘要:

Audio is the main form for the visually impaired to obtain information. In reality, all kinds of visual data always exist, but audio data does not exist in many cases. In order to help the visually impaired people to better perceive the information around them, an image-to-audio-description (I2AD) task is proposed to generate audio descriptions from images in this paper. To complete this totally new task, a modal translation network (MT-Net) from visual to auditory sense is proposed. The proposed MT-Net includes three progressive sub-networks: 1) feature learning, 2) cross-modal mapping, and 3) audio generation. First, the feature learning sub-network aims to learn semantic features from image and audio, including image feature learning and audio feature learning. Second, the cross-modal mapping sub-network transforms the image feature into a cross-modal representation with the same semantic concept as the audio feature. In this way, the correlation of inter-modal data is effectively mined for easing the heterogeneous gap between image and audio. Finally, the audio generation sub-network is designed to generate the audio waveform from the cross-modal representation. The generated audio waveform is interpolated to obtain the corresponding audio file according to the sample frequency. Being the first attempt to explore the I2AD task, three large-scale datasets with plenty of manual audio descriptions are built. Experiments on the datasets verify the feasibility of generating intelligible audio from an image directly and the effectiveness of proposed method.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑