Visual Dialog requires an AI agent to hold a meaningful dialogue with humans in natural, conversational language about visual content. Specifically, given an image, a dialogue history, and a follow-up question about the image, the task is to answer the question.

Dataset stats:
120k images from COCO
1 dialog / image
10 rounds of question-answers /dialogue
Total 1.2M dialogue question-answers Abhishek Das, Satwik KotturDeshraj YadavPrithvijit ChattopadhyayViraj PrabhuArjun ChandrasekaranNirbhay ModheKhushi GuptaAvi SinghJosé M. F. MouraStefan LeeDevi Parikh, Dhruv Batra

< Prev Next >