Digital cameras and sensors are used increasingly in a range of monitoring or exploratory applications, in particular for biological, geological and physical surveys. These technologies have revolutionized our ability to capture multimedia data on large scale for environmental monitoring and are also greatly improving our ability to effectively manage natural resources and increasing our competitiveness. For example, the Xeno-canto  project has recorded several thousands of bird sounds from all over the world; the Pl@ntNet project has collected massive plant images for analyzing geographic distribution of plants in the Mediterranean area; the Fish4Knowledge project has recorded many Terabytes of video data for monitoring Taiwanese coral reefs.

Despite the recent technology advances has enabled massive data collection, its analysis usually requires very time-consuming and expensive input by human observers. This analytical “bottleneck” greatly restricts the use of these technologies and demands for efficient organization, browsing and retrieval tools to enable proactive provision of analytical information.

The automated analysis of such ecology-related multimedia data presents new challenges and  the results are of great interest to the general public (e.g. considering mobile-based plant recognition applications for amateurs) as well as to domain experts. Examples of the latter include biologists working on understanding the natural environment, promoting its preservation, and studying the behavior and interactions of the living organisms (insects, animals, etc.) that are part of it.

This workshop features two tracks:

  • Regular Track aiming to present and report on the most recent methods for the management, processing, interpretation, and visualization of multimedia data recorded for monitoring ecological systems, with particular attention to the understanding of the behavior of animals and insects.
  • Data Challenge Track aiming to test image retrieval, computer vision and machine learning methods on a complex and extensive ecological visual dataset, which poses significant challenges in large-scale visual analysis, feature indexing and similarity search.

Papers may be submitted to any of the two tracks, and there is no requirement for participating to the data challenge in order to submit a regular paper, and vice versa. Both regular and data challenge submissions will be peer-reviewed and accepted submissions will be included in the workshop proceedings.