Advancements in Face Sketch Recognition: Techniques and Challenges

Face is a key contributor for criminal identification in law enforcement agencies. Forensic Sketches act as a link between suspect identification and eyewitness descriptions, which is essential to criminal investigations. The sketch artist in forensic investigations typically develop their sketches based on spoken information given by an eyewitness regarding the suspect. These drawings are created by the artist either by hand (Forensic sketches) or with the aid of a tool (Composite sketches). There is dissonance in the facial features of a sketch since the eyewitness’s memory is used to describe the features of the face. Traditional approaches relied heavily on manual comparison, which is time-consuming and prone to human error. Advances in the field of Machine Learning and Computer Vision have led to considerable improvements in the accuracy and efficiency of matching forensic/composite sketching with digital photos. Recently, A range of sketch-to-photograph retrieval strategies have been presented by various researchers, nevertheless they often ignore the uncertainty of facial characteristics for questionable faces. This corresponding study gives a thorough analysis of the approaches which are currently available for matching suspect face sketches with digital photos. The review focuses on the dataset, conventional techniques and its classification and demonstrates how suspect face retrieval methods have changed over time. Additionally, it provides a summary of the overall effectiveness of excellent state-of-the-art methods.