日本画像学会誌
Online ISSN : 1880-4675
Print ISSN : 1344-4425
ISSN-L : 1344-4425
Special Topic
Generating Easy-to-understand Referring Expressions for Target Identifications
Mikihiro TANAKATakayuki ITAMOCHIKenichi NARIOKAIkuro SATOYoshitaka USHIKUTatsuya HARADA
著者情報
ジャーナル フリー

2020 年 59 巻 6 号 p. 591-600

詳細
抄録

For communication between humans and intelligent agents such as robots, it is an important issue for agents to tell humans what they see. In this article, we introduce the results of our research on the generation of sentences that not only refer to objects correctly but also let humans find them quickly. If the target is not salient, finding the target itself becomes difficult. Therefore, we designed the model to utilize the salient contexts around it (e.g. “beside a car”) to help humans to find the targets. Moreover, we optimized the generation of sentences that are easily understood by using the time required to locate the referred objects by humans and their accuracies. To evaluate our system, we created a new dataset using images from Grand Theft Auto V (GTA V). Experimental results showed that our system generated sentences that are easily comprehended by humans, especially for less salient targets.

著者関連情報
© 2020 by The Imaging Society of Japan
前の記事 次の記事
feedback
Top