English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Near field depth estimation around a self driving car is an importantfunction that can be achieved by four wide angle fisheye cameras having a fieldof view of over 180. Depth estimation based on convolutional neural networks(CNNs) produce state of the art results, but progress is hindered because depthannotation cannot be obtained manually. Synthetic datasets are commonly usedbut they have limitations. For instance, they do not capture the extensivevariability in the appearance of objects like vehicles present in realdatasets. There is also a domain shift while performing inference on naturalimages illustrated by many attempts to handle the domain adaptation explicitly.In this work, we explore an alternate approach of training using sparse LiDARdata as ground truth for depth estimation for fisheye camera. We built our owndataset using our self driving car setup which has a 64 beam Velodyne LiDAR andfour wide angle fisheye cameras. To handle the difference in view points ofLiDAR and fisheye camera, an occlusion resolution mechanism was implemented. Westarted with Eigen's multiscale convolutional network architecture and improvedby modifying activation function and optimizer. We obtained promising resultson our dataset with RMSE errors comparable to the state of the art resultsobtained on KITTI.

Monocular Fisheye Camera Depth Estimation Using Sparse LiDAR Supervision