paper-with-me

SQA3D

Situated Question Answering in 3D Scenes

홈페이지 · 논문 58편

SQA3D is a dataset for embodied scene understanding, where an agent needs to comprehend the scene it situates from an first person's perspective and answer questions. The questions are designed to be situated, embodied and knowledge-intensive. We offer three different modalities to represent a 3D scene: 3D scan, egocentric video and BEV picture.

ImagesVideosTexts3D English

벤치마크

3D Question Answering (3D-QA) on SQA3D 결과 13개
Question Answering on SQA3D 결과 8개
Referring Expression on SQA3D 결과 1개