본문 바로 가기

로고

국내 최대 기계 및 로봇 연구정보
통합검색 화살표
  • 유니맷 베이직에듀
  • 우수 논문

    뷰페이지 제목 게시판 내용

    Text2action: Generative adversarial synthesis from language to action  
    Ahn, Hyemin(Seoul National University)
    Australia | ICRA 2018
    2018-05-21 | 바로가기
    Decoding, Robots, Gallium_nitride
    Cited by 24

    ■  View full text

    2018 IEEE International Conference on Robotics and Automation (ICRA)

    Date of Conference: 21-25 May 2018

    https://doi.org/10.1109/ICRA.2018.8460608

     

    ■  Researchers

    Hyemin Ahn, Timothy Ha, Yunho Choi, Hwiyeon Yoo, Songhwai Oh

    Department of Electrical and Computer Engineering and ASRI, Seoul National Uni-versity

     

     

    ■  Abstract

    In this paper, we propose a generative model which learns the relationship between language and human action in order to generate a human action sequence given a sentence describing human behavior. The proposed generative model is a generative adversarial network (GAN), which is based on the sequence to sequence (SEQ2SEQ) model. Using the proposed generative network, we can synthesize various actions for a robot or a virtual agent using a text encoder recurrent neural network (RNN) and an action decoder RNN. The proposed generative network is trained from 29,770 pairs of actions and sentence annotations extracted from MSR-Video-to-Text (MSR-VTT), a large-scale video dataset. We demonstrate that the network can generate human-like actions which can be transferred to a Baxter robot, such that the robot performs an action based on a provided sentence. Results show that the proposed generative network correctly models the relationship between language and action and can generate a diverse set of actions from the same sentence.

     

     

    ■  Conclusion

    In this paper, we have proposed a generative model based on the Seq2seq model [7] and generative adversarial network (GAN) [8], for enabling a robot to execute various actions corresponding to an input language deion. It is interesting to note that our generative model, which is different from other existing related works in terms of utilizing the advantages of the GAN, is able to generate diverse behaviors when the input random vector sequence changes. In addition, results show that our network can generate an action sequence that is more dynamic and closer to the actual data than the network presented presented in [5]. The proposed generative model, which understands the relationship between the human language and the action, generates an action corresponding to the input language. We believe that the proposed method can make actions by robots more understandable to their users.

     

     

     

     

    * 관련 자료 

    [신진연구자 인터뷰] 이미지 및 언어 인지에 기반한 로봇 행동 생성을 위한 학습 방법론 연구

    전체댓글0

    댓글 입력란
    프로필 이미지
    0/250자

    서브 사이드

    서브 우측상단1