AU Dataset for Visuo-Haptic Object Recognition for Robots
收藏资源简介:
AU数据集是专为机器人设计的视觉-触觉物体识别数据集,由奥胡斯大学电气与计算机工程系创建。该数据集包含63个物体的视觉、动觉和触觉(音频/振动)数据,旨在解决多模态物体识别中的视觉和触觉模糊性问题。数据集创建过程中,采用了多种探索程序来捕捉物体的视觉和触觉特性,如视觉探索、动觉探索和触觉探索。该数据集适用于机器人感知研究,特别是多模态物体识别和数据融合领域,旨在通过整合视觉和触觉信息提高物体识别的准确性。
The AU Dataset is a visual-tactile object recognition dataset designed specifically for robotics, created by the Department of Electrical and Computer Engineering at Aarhus University. This dataset contains visual, kinesthetic, and tactile (audio/vibration) data for 63 distinct objects, aiming to address the visual and tactile ambiguities in multimodal object recognition. During the dataset creation process, multiple exploration procedures were adopted to capture the visual and tactile properties of the objects, including visual exploration, kinesthetic exploration and tactile exploration. This dataset is applicable to robotics perception research, particularly in the fields of multimodal object recognition and data fusion, with the objective of improving the accuracy of object recognition by integrating visual and tactile information.

- 1AU Dataset for Visuo-Haptic Object Recognition for Robots奥胡斯大学电气与计算机工程系 · 2021年



