iBims-1 高质量 RGB-D 数据集

Ibims-1 全称 Independent benchmark images and matched scans v1，是一个新的高质量 RGB-D 数据集。其专为测试单图像深度估计（SIDE）方法而设计。研究人员用数字单镜头反射（DSLR）相机和高精度激光扫描仪组成的定制采集设置，获取不同室内场景的高分辨率图像和高精度深度图。与其他 RGB-D 数据集相比，iBims-1 数据集噪音水平低、深度转换急剧、无遮挡和深度范围高的优势显著。

Ibims-1, whose full name is Independent Benchmark Images and Matched Scans v1, is a novel high-quality RGB-D dataset specifically designed for testing single-image depth estimation (SIDE) methods. Researchers employed a custom acquisition setup consisting of a digital single-lens reflex (DSLR) camera and a high-precision laser scanner to capture high-resolution RGB images and high-precision depth maps of various indoor scenes. Compared with other RGB-D datasets, the iBims-1 dataset boasts prominent advantages including low noise levels, sharp depth transitions, no occlusion issues, and a wide depth range.

搜集汇总

背景与挑战

背景概述

iBims-1是一个专为单图像深度估计方法测试设计的高质量RGB-D数据集，通过定制采集设置获取了低噪音、高精度的室内场景图像和深度图。该数据集包含100对高分辨率RGB-D图像、多种遮罩及校准参数，并提供了颜色几何增强、手持图像等辅助数据。