From A Glance to "Gotcha": Interactive Facial Image Retrieval with Progressive Relevance Feedback
This work addresses the challenge for untrained witnesses in forensic investigations to identify suspects from large image pools, though it is incremental as it builds on existing relevance feedback methods.
The paper tackles the problem of facial image retrieval for forensic investigations by proposing an end-to-end framework that uses progressive relevance feedback from witnesses, achieving 99% ranking percentile performance on the CelebA dataset.
Facial image retrieval plays a significant role in forensic investigations where an untrained witness tries to identify a suspect from a massive pool of images. However, due to the difficulties in describing human facial appearances verbally and directly, people naturally tend to depict by referring to well-known existing images and comparing specific areas of faces with them and it is also challenging to provide complete comparison at each time. Therefore, we propose an end-to-end framework to retrieve facial images with relevance feedback progressively provided by the witness, enabling an exploitation of history information during multiple rounds and an interactive and iterative approach to retrieving the mental image. With no need of any extra annotations, our model can be applied at the cost of a little response effort. We experiment on \texttt{CelebA} and evaluate the performance by ranking percentile and achieve 99\% under the best setting. Since this topic remains little explored to the best of our knowledge, we hope our work can serve as a stepping stone for further research.