Italian Trulli

Introduction

Depth estimation is a core task in 3D computer vision. Our dataset, HAMMER is a multi-modal dataset comprises of depth maps from multiple commonly used sensors for indoor depth estimation, namely LiDAR, ToF, stereo, active stereo together with monocular RGB+P data as well as highly reliable ground truth depth maps via depth rendering of annotated 3D meshes of objects and backgrounds scanned by 3D scanners and camera pose provided by robotic arm. Depth estimators can be trained on this data with RGB and any combination of typical depth sensors with the provided reliable ground truth. Our dataset HAMMER, which we make publicly available can provide a reliable base to pave the way to targeted depth improvements and sensor fusion approaches.

Format

The dataset contains 13 Scenes. Each scene has two setups, with/without objects (with : scene includes several objects with various surface material, without : scene with only backgrounds - naked) and each scene has two camera trajectories. Each trajectories composed with roughly 300 frames, which adds up to 16k frames in total (13 x 2 x 2 x 300). Each trajectory contains corresponding images from each cameras : d435 – stereo, l515 – Lidar (D-ToF), polarization – RGBP (RGB with polarization), tof – (I-ToF). Each camera folder contains its intrinsic file and its own recorded images together with rendered depth GT / instance GT and camera pose. All the cameras are fully synchronized via robotic arm’s data acquisition setup.

Download

The dataset is published under MIT license and is free to use for commercial purposes. Link