Original Website: https://sites.google.com/a/6d-vision.com/www/current-research/lostandfounddataset
The LostAndFound Dataset addresses the problem of detecting unexpected small obstacles on the road often caused by lost cargo.
The dataset comprises 112 stereo video sequences with 2104 annotated frames (picking roughly every tenth frame from the recorded data).
If you are using this dataset in a publication please cite the following paper:
Peter Pinggera, Sebastian Ramos, Stefan Gehrig, Uwe Franke, Carsten Rother, Rudolf Mester, "Lost and Found: Detecting Small Road Hazards for Self-Driving Vehicles", Proceedings of IROS 2016, Daejeon, Korea. Link to the paper
(This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.)
For the data format and the interpretation of the data sources we refer to the description of the Cityscapes dataset format which we closely follow: http://www.cityscapes-dataset.com
Below you can find a link to the data description and some development kit (tailored for Cityscapes but applicable to LostAndFound as well):
In order to replace the cityscapes mapping with lostAndFound labels replace labels.py in the development kit with this file: labels.py
A description of the labels of the LostAndFound dataset can be found here: laf_table.pdf
Below, you can find all currently available downloads. A README and various scripts for inspection, preparation, and evaluation can be found in above git repository.
The following packages are available for download:
gtCoarse.zip (37MB) annotations for train and test sets (2104 annotated images)
leftImg8bit.zip (6GB) left 8-bit images - train and test set (2104 images)
rightImg8bit.zip (6GB) right 8-bit images - train and test set (2104 images)
leftImg16bit.zip (17GB) right 16-bit images - train and test set (2104 images)
rightImg16bit.zip (17GB) right 16-bit images - train and test set (2104 images)
disparity.zip (1.4GB) depth maps using Semi-Global Matching for train and test set (2104 images)
timestamp.tgz (50kB) timestamps for train and test sets
camera.zip (1MB) Intrinsic and extrinsic camera parameters for train and test sets
vehicle.zip (1MB) vehicle odometry data (speed and yaw rate) for train and test sets
The LostAndFound dataset may be used according to the following license agreement:
---------------------- The LostAndFound Dataset ----------------------
This dataset is made freely available to academic and non-academic entities for non-commercial purposes such as academic research, teaching, scientific publications, or personal experimentation. Permission is granted to use the data given that you agree:
1. That the dataset comes "AS IS", without express or implied warranty. Although every effort has been made to ensure accuracy, we (Daimler AG) do not accept any responsibility for errors or omissions.
2. That you include a reference to the LostAndFound Dataset in any work that makes use of the dataset. For research papers, cite our preferred publication as listed on our website; for other media link to the dataset website.
3. That you do not distribute this dataset or modified versions. It is permissible to distribute derivative works in as far as they are abstract representations of this dataset (such as machine learning models trained on it or additional annotations that do not directly include any of our data) and do not allow to recover the dataset or something similar in character.
4. That you may not use the dataset or any derivative work for commercial purposes as, for example, licensing or selling the data, or using the data with a purpose to procure a commercial gain. 5. That all rights not expressly granted to you are reserved by us (Daimler AG).
Sebastian Ramos, Peter Pinggera, Stefan Gehrig
For questions, suggestions, and comments contact Stefan Gehrig (Stefan.Gehrig (at) daimler.com) or Sebastian Ramos.