Headcam: Cylindrical Panoramic Video Dataset for Unsupervised Learning of Depth and Ego-Motion

<p>This dataset contains panoramic video captured from a helmet-mounted camera while riding a bike through suburban Northern Virginia. &nbsp;We used the videos to evaluate an unsupervised learning method for depth and ego-motion&nbsp;estimation, as described in our paper:</p> <p>Alisha Sharma and Jonathan Ventura. &nbsp;&quot;Unsupervised Learning of Depth and Ego-Motion from Cylindrical Panoramic Video.&quot; &nbsp;Proceedings of the 2019 IEEE Artificial Intelligence &amp; Virtual Reality Conference, San Diego, CA, 2019.</p> <p>If you make use of this dataset, please cite this paper.</p> <p>&nbsp;</p> <p>The videos are stored as .mkv video files encoded using lossless H.264. &nbsp;To extract the images, we recommend using ffmpeg:</p> <blockquote> <p>mkdir 2018-10-03 ;</p> <p>ffmpeg -i&nbsp;2018-10-03.mkv -q:v 1 2018-10-03/%05d.png ;</p> </blockquote> <p>Associated code can be found in our <a href="https://github.com/jonathanventura/cylindricalsfmlearner">GitHub repository</a>. &nbsp;</p>