Overview

We provide a new synthetic dataset called SYNTH3 specifically developed for machine learning based Stereo-ToF fusion applications. The dataset is split in two parts, a training set and a test set. The training set contains 40 scenes obtained by rendering 20 unique scenes from different viewpoints while the test set is composed by 15 unique scenes. The various scenes contain furnitures and objects of various shapes in different environments e.g., living rooms, kitchen rooms or offices. Furthermore, some outdoor locations with non-regular structure are also included in the dataset.They appear realistic and suitable for the simulation of Stereo-ToF acquisition systems

Data Format

For each scene sample in the dataset, the following data are provided: (NEW: the dataset has been updated and extended and now contains also reprojected data and the ToF acquisition at different frequencies)

• The 512x424 ToF depth map.
• The 960x540 ToF depth map projected on the reference camera of the stereo system.
• The 512x424 ToF amplitude images captured at 16, 80 and 120 MHz respectively.
• The 960x540 ToF amplitude image captured at 120 MHz and projected on the reference camera of the stereo system.
• The 512x424 ToF intensity images captured at 16, 80 and 120 MHz respectively.
• The ground truth depth map w.r.t. the ToF point of view.
• The 1920x1080 color image acquired by the left camera of the stereo system.
• The 1920x1080 color image acquired by the right camera of the stereo system.
• The 960x540 disparity and depth maps estimated from the color images of the stereo system w.r.t. the right camera.
• The ground truth depth and disparity maps w.r.t. the right camera of the stereo system.

Scene samples from the test set, color images acquired by the right camera of the stereo system (images at a lower resolution for display purposes only)

Citation

Please use the following citation when referencing the dataset:

@INPROCEEDINGS{8265297,
  author={G. {Agresti} and L. {Minto} and G. {Marin} and P. {Zanuttigh}},
  booktitle={2017 IEEE International Conference on Computer Vision Workshops (ICCVW)},
  title={Deep Learning for Confidence Information in Stereo and ToF Data Fusion},
  year={2017},
  volume={},
  number={},
  pages={697-705},}