A VR Interface for Browsing Visual Spaces at VBS2021

Abstract

The Video Browser Showdown (VBS) is an annual competition in which each participant prepares an interactive video retrieval system and partakes in a live comparative evaluation at the annual MMM Conference. In this paper, we introduce Eolas, which is a prototype video/image retrieval system incorporating a novel virtual reality (VR) interface. For VBS'21, Eolas represented each keyframe of the collection by an embedded feature in a latent vector space, into which a query would also be projected to facilitate retrieval within a VR environment. A user could then explore the space and perform one of a number of filter operations to traverse the space and locate the correct result.

Videofall - A Hierarchical Search Engine for VBS2022

“…Then, the result was presented in a 2D scrolling list according to the ranked list of scenes. Different from vitrivr-VR, the main goal of EOLAS [8] -another VR system first introduced last yearwas to explore the latent feature space where all frames were embedded. Based on the cosine distance between encoded vectors, the group of videos would be visualized on the wide and surrounding 3D view in the virtual environment.…”

Multi-modal Video Retrieval in Virtual Reality with vitrivr-VR

“…These advantages are explored by vitrivr-VR, which combines a VR-based user interface with the vitrivr stack 3 , an open-source multimedia retrieval system. One way to evaluate the effectiveness of novel retrieval systems is through evaluation campaigns like the Video Browser Showdown (VBS) [10], in which other VR systems, such as EOLAS [16], also participate. This paper describes the version of vitrivr-VR with which we plan to participate in the VBS 2022.…”