CoSy logo Cognitive Systems for Cognitive Assistants

CoSy Book Datasets

KTH-IDOL2 Database - Image Database for rObot Localization
The name IDOL is an acronym which stands for Image Database for rObot Localization. The database consists of 24 image sequences accompanied by laser scans and odometry data acquired using two mobile robot platforms. The acquisition was performed within an indoor laboratory environment consisting of five rooms of different functionality (one-person office, two-persons office, corridor, kitchen, and printer area) under various illumination conditions (in cloudy weather, in sunny weather, and at night) across a span of 6 months. As a result, the data capture natural variability that occur in real-world environments introduced by both illumination and human activity. The KTH-IDOL2 database is an extension of the KTH-IDOL1 database and as such consists of 12 sequences taken from the KTH-IDOL1 database and another 12 sequences acquired 6 months later.

INDECS Database - Indoor Environment under Changing conditionS
The name INDECS is an acronym which stands for Indoor Environment under Changing conditionS. The database consists of several sets of pictures taken in five rooms of different functionality under various illumination and weather conditions at different periods of time. Each room was observed from many viewpoints and angles. Moreover, the normal activity in the rooms was recorded: people appear in the rooms, pieces of furniture are moved over time.

COLD Database - CoSy Localization Database
The name COLD is an acronym which stands for COsy Localization Database. The database represents an effort to provide a large-scale, flexible testing environment for evaluating mainly vision-based localization systems aiming to work on mobile platforms in realistic settings. The COLD database consists of three separate datasets acquired at three different indoor laboratory environments located in three different European cities. The database contains image sequences captured using a regular and omni-directional cameras together with laser range scans and odometry data. The data were recorded using three different mobile robot platforms and the same camera setup, under various weather and illumination conditions (during cloudy weather, sunny weather and at night) over several days. In each of the three labs, the acquisition was performed in several rooms of different functionality. Consequently, the COLD database is an ideal testbed for assessing the robustness of localization and place recognition/categorization algorithms with respect to both categorical and dynamic changes (introduced by illumination variations and human activity).


Last modified: 7.1.2009 17:23:27