Master Assignment
Robust Tattoo Detection and Retrieval
Type: Master EE/CS
Period: TBD
Student: Unassigned
If you are interested please contact :
Introduction:
Despite the enormous progress in biometrics-based primary modalities such as the face, iris and fingerprint, unimodal biometrics identification has not been accepted in forensics. Tattoos, which constitute a pertinent and highly distinctive soft biometric trait, can be particularly useful in describing wanted or missing people, or even unidentied bodies. Hence, tattoos are highly instrumental in person identification. Consequently, research on tattoobased biometrics has gained signicant interest in the last few years. In order to explore the potential of tattoos, various research directions have been proposed in the literature. However, existing tattoo search methods or tattoo retrieval techniques mainly focus on the matching of cropped tattoos. Therefore, these topics require more analysis. Open research problems include tattoo detection, as well as localization, i.e., determining whether an image contains a tattoo and if so, segmentation of the tattoo.
Figure 1: Examples of tattoo images and sketches
Assignment:
Benchmark Dataset: The purpose of the competition is to benchmark tattoo detection, tattoo retrieval and tattoo sketch-based retrieval with 3 tattoo datasets. For the tattoo detection task, two different datasets will be employed. The first dataset Web Tattoo is collected from the Internet, and the labelling team has annotated the bounding boxes. The second dataset is the public dataset which was used in Tatt-C competition. For the tattoo retrieval task, we have a tattoo retrieval dataset that includes 399 training instances and 200 test instances. For the tattoo sketch retrieval task the dataset was prepared by drawing 300 tattoo sketches by volunteers, who was asked to take a look at a tattoo image for one minute and then draw the tattoo sketch the next day (see an example of the tattoo sketch. This assignment is linked to the Tattoo Detection and Retreival competition if the BTAS 2019 conference. More information and the data can be obtained here: https://sites.google.com/site/rtdrc2019/