Introduction to the Third Annual Lifelog Search Challenge (LSC'20)

Abstract

The Lifelog Search Challenge (LSC) is an annual comparative benchmarking activity for comparing approaches to interactive retrieval from multi-modal lifelogs. LSC'20, the third such challenge, attracts fourteen participants with their interactive lifelog retrieval systems. These systems are comparatively evaluated in front of a live-audience at the LSC workshop at ACM ICMR'20 in Dublin, Ireland. This overview motivates the challenge, presents the dataset and system configuration used in the challenge, and briefly presents the participating teams. CCS CONCEPTS • Human-centered computing → Empirical studies in interaction design; • Information systems → Mobile information processing systems; Search interfaces.

FIRST - Flexible Interactive Retrieval SysTem for Visual Lifelog Exploration at LSC 2020

“…In the recent ImageCLEF lifelog task [1] and the recent NTCIR14-Lifelog task [4], the retrieval tools are used by their creators who have in-depth knowledge about their systems, and the objective for participants mainly focuses on how to better extract information from visual lifelog data. In Lifelog Search Challenge (LSC [5]), the goals are not only to improve image understanding but also to design and enhance usability for professional and novice users to interact with the retrieval systems to handle various query types.…”

“…Lifelog [5] data provides valuable information for people to analyze their daily activities, events, and memories. However, it is not easy to describe what we want to search for from a massive amount of lifelog data, especially when much of the data is in a visual format.…”

“…The annual Lifelog Search Challenge (LSC [5]) aims to evaluate different approaches to assist users seeking to find events/moments of interest in an interactive manner from lifelog data. Many systems have been developed with different methods of image analysis and understanding, and different modalities for expressing queries and user interaction for result refinement [6].…”

LifeGraph: A Knowledge Graph for Lifelogs

“…In order to maximize intelligence, we fuse information from multiple sources that will be briefly described. Naturally, our starting point is the core image data set provided by the organizers (and obtainable from the LSC website) [7]. In the following, we will explain how we linked it to two additional data sources which will be described here first: the "Classification of Everyday Living" (COEL) [3] and Wikidata [26].…”

Myscéal: An Experimental Interactive Lifelog Retrieval System for LSC'20

“…Different from the previous LSC competitions, LSC'20 uses the personal data of a single lifelogger over four months between the years 2015 to 2018 [12]. Images in the dataset are redacted by blurring faces and sensitive textual information.…”