Composite Method

This paper explores how race comes to matter in the practice of police facial composite drawing. The confidential nature of criminal investigations prevented us from using research material collected through observations of police practices. The authors developed an experimental film project in collaboration with two forensic artists to illuminate the production of (visual) differences in the context of facial composite drawing. We recorded the process using a variety of technologies to produce different materializations of the drawing event. The experimental setting created a reflexive space for all participants, albeit not in the same way. Tinkering with the materials generated allowed us to analyze the enactment and slipperiness of race in practice. This paper combines written text with experimental montage to address three different practices through which race takes shape in the process of making facial composite drawings: 1) touching as describing; 2) layering and surfacing; and 3) articulating the common.

CG-GAN: An Interactive Evolutionary GAN-Based Approach for Facial Composite Generation

Facial composites are graphical representations of an eyewitness's memory of a face. Many digital systems are available for the creation of such composites but are either unable to reproduce features unless previously designed or do not allow holistic changes to the image. In this paper, we improve the efficiency of composite creation by removing the reliance on expert knowledge and letting the system learn to represent faces from examples. The novel approach, Composite Generating GAN (CG-GAN), applies generative and evolutionary computation to allow casual users to easily create facial composites. Specifically, CG-GAN utilizes the generator network of a pg-GAN to create high-resolution human faces. Users are provided with several functions to interactively breed and edit faces. CG-GAN offers a novel way of generating and handling static and animated photo-realistic facial composites, with the possibility of combining multiple representations of the same perpetrator, generated by different eyewitnesses.

Confirmation bias in decision making for fingerprints, DNA and eyewitness evidence cannot be explained by cognitive style or thinking dispositions.

Pre-existing beliefs about the background or guilt of a suspect can bias the subsequent evaluation of evidence for forensic examiners and lay people alike. This biasing effect, called the confirmation bias, has influenced legal proceedings in prominent court cases such as that of Brandon Mayfield. Today many forensic providers attempt to train their examiners against these cognitive biases. Nine hundred and forty-two participants read a fictional criminal case and received either neutral, incriminating or exonerating evidence (fingerprint, eyewitness, or DNA) before providing an initial rating of guilt. Participants then viewed ambiguous evidence (alibi, facial composite, handwriting sample or informant statement) before providing a final rating of guilt. Final guilt ratings were higher for all evidence conditions (neutral, incriminating or exonerating) following exposure to the ambiguous evidence. This provides evidence that the confirmation bias influences the evaluation of evidence.