Dataset for "Object Detection in Equirectangular Panorama"
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Panoramic Object Detection Dataset


This repository contains following contents:

  • the dataset collected and used in "Object Detection in Equirectangular Panorama"
  • the evaluation code (to be added)
  • the comparison experiment results
  • the annotation tool


Object Detection in Equirectangular Panorama

Wenyan Yang*, Yanlin Qian*, Joni-Kristian Kämäräinen*, Francesco Cricri*, Lixin Fan*

International Conference on Pattern Recognition (ICPR) 2018


Downloads: link Slides: link

It containes three types of the ground truth: BFOV (Bounding FOV), BBox(Bounding box) and 3D-Container frame.

  • BFOV (path: 360-Dataset/bbx-annotations/*)
  • BFOV (path: 360-Dataset/bfov-annotations/*)
	<center_lon> // centre's longitude
	<center_lat> // centre's latitude
	<bfov_w> // the width's radius
	<bfov_h> //  the height's radius
  • 3D-Container Frame(path: 360-Dataset/bfov-annotations/*)
// The 4 vertexes and 1 centre point of the BFOV. Stored in radians
      <tl_lon> // top left vertex's longitude.
      <tl_lat> // top left vertex's latitude.
      <tr_lon> // top right vertex's longitude.
      <tr_lat> // top right vertex's latitude.
      <br_lon> // bottom right vertex's longitude.
      <br_lat> // bottom right vertex's latitude.
      <bl_lon> // bottom left vertex's longitude.
      <bl_lat> // bottom left vertex's latitude.
      <c_lon> // centre's longitude
      <c_lat> // centre's latitude.


Youtube: link

  • To visualize the BBOX annotation, some traditional annotation tools are needed (e.g:
  • To visualize the BFOV annotation, use the tool in folder "/visualize_tool" (open the web browser and load the image and XML file)

Annotation tool

To be added

Comparison experiments results

To be added

Running the tests

To be added


This project is licensed under the MIT License - see the file for details


  • Please cite our paper if you find this work is useful.
author={W. Yang and Y. Qian and J. Kämäräinen and F. Cricri and L. Fan}, 
	booktitle={2018 24th International Conference on Pattern Recognition (ICPR)}, 
	title={Object Detection in Equirectangular Panorama}, 
	keywords={Detectors;Distortion;Object detection;Cameras;Image resolution;Virtual reality;Graphics processing units},