装配特征点识别数据集

该数据集采用Intel RealSense D435i相机进行图像数据采集，并使用光流法完成特征点数据的半自动标注。数据集包含D-Sub和RJ45两种3C场景中常见的连接器，共采集了500组高质量标注的数据。该数据集为装配件识别、装配语义提取和机器人灵巧操作技能学习研究提供了重要的数据支撑，具有显著的科学价值和应用前景。

This dataset was collected using an Intel RealSense D435i camera for image data acquisition, and semi-automated annotation of feature point data was performed via the optical flow method. The dataset encompasses two types of common connectors in 3C scenarios, namely D-Sub and RJ45, with a total of 500 sets of high-quality annotated data collected. This dataset provides vital data support for research on component recognition, assembly semantic extraction, and robotic dexterous manipulation skill learning, and holds considerable scientific value and promising application prospects.

搜集汇总

背景与挑战

背景概述

该数据集采用Intel RealSense D435i相机采集图像，并利用光流法实现了特征点数据的半自动标注。它包含D-Sub和RJ45两种3C场景常见连接器的500组高质量标注数据，为装配件识别、装配语义提取和机器人灵巧操作技能学习提供了重要的数据支撑。