申请试用
HOT
登录
注册
 
LiveBot: Generating Live Video Comments Based on Visual and Textual Contexts

LiveBot: Generating Live Video Comments Based on Visual and Textual Contexts

Reboot
/
发布于
/
2399
人观看
We introduce the task of automatic live commenting. Live commenting, which is also called “video barrage”, is an emerging feature on online video sites that allows real-time comments from viewers to fly across the screen like bullets or roll at the right side of the screen. The live comments are a mixture of opinions for the video and the chit chats with other comments. Automatic live commenting requires AI agents to comprehend the videos and interact with human viewers who also make the comments, so it is a good testbed of an AI agent’s ability to deal with both dynamic vision and language. In this work, we construct a large-scale live comment dataset with 2,361 videos and 895,929 live comments. Then, we introduce two neural models to generate live comments based on the visual and textual contexts,which achieve better performance than previous neural base lines such as the sequence-to-sequence model. Finally, we provide a retrieval-based evaluation protocol for automatic live commenting where the model is asked to sort a set of candidate comments based on the log-likelihood score, and evaluated on metrics such as mean-reciprocal-rank. Putting it all together, we demonstrate the first “LiveBot”. The datasets and the codes can be found at https://github.com/lancopku/livebot.
6点赞
0收藏
0下载
相关推荐
确认
3秒后跳转登录页面
去登陆