SHOP-VRB
收藏资源简介:
SHOP-VRB是一个专注于视觉推理的基准数据集,特别适用于机器人应用中的小对象抓取和操作。该数据集由帝国理工学院电气与电子工程系创建,包含66个3D模型,涵盖多种家庭物品及其属性、功能和自然语言描述。数据集通过合成数据生成方法扩展,支持更复杂的视觉问题回答和场景理解。SHOP-VRB旨在解决现有数据集在视觉推理和机器人感知任务中的局限性,提供更接近真实世界应用的挑战性测试环境。
SHOP-VRB is a benchmark dataset focused on visual reasoning, specifically tailored for small object grasping and manipulation in robotic applications. Developed by the Department of Electrical and Electronic Engineering at Imperial College London, this dataset contains 66 3D models covering a variety of household items, along with their attributes, functions, and natural language descriptions. The dataset is expanded via synthetic data generation methods, enabling more complex visual question answering and scene understanding tasks. SHOP-VRB aims to address the limitations of existing datasets in visual reasoning and robotic perception tasks, providing a challenging test environment that is more aligned with real-world application scenarios.

- 1SHOP-VRB: A Visual Reasoning Benchmark for Object Perception帝国理工学院电气与电子工程系 · 2020年



