Virtual KITTI

Virtual KITTI is a photo-realistic synthetic video dataset designed to learn and evaluate computer vision models for several video understanding tasks: object detection and multi-object tracking, scene-level and instance-level semantic segmentation, optical flow, and depth estimation.

Released in: Virtual KITTI

Source: arXiv - Virtual Worlds as Proxy for Multi-Object Tracking Analysis

Contributor:

by

Homepage

Summary

Virtual KITTI is a photo-realistic synthetic video dataset designed to learn and evaluate computer vision models for several video understanding tasks: object detection and multi-object tracking, scene-level and instance-level semantic segmentation, optical flow, and depth estimation.

21260

Images in dataset

2016

Year Released

Key Links & Stats

VisualComputingInstitute / vkitti3D-dataset

@inproceedings{Gaidon:Virtual:CVPR2016, author = {Gaidon, A and Wang, Q and Cabon, Y and Vig, E}, title = {Virtual Worlds as Proxy for Multi-Object Tracking Analysis}, booktitle = {CVPR}, year = {2016} }

scenebox

Modalities

  1. Video

Verticals

  1. A/V

ML Task

  1. Image Classification
  2. Object Detection
  3. Autonomous Vehicles

Related organizations