Facial composite systems: review

Zahradníková, Barbora; Duchovičová, Soňa; Schreiber, Peter

doi:10.1007/s10462-016-9519-1

Cited by 11 publications

(13 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The objective evaluation of the system is a complex task as it depends on subjective recognition abilities (Zahradníková, Duchovičová, and Schreiber 2016). To overcome this issue we use two separated evaluation metrics:…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…Recall is the recollection of information for the composite construction. Recognition is the ability to recognize someone seen before, used during line-ups (Zahradníková, Duchovičová, and Schreiber 2016;Mancusi 2014). Compared to recall, recognition is an easier and stronger task, therefore most courts of law consider lineup identification as good evidence (Mancusi 2014).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Zaltron

Zurlo

Risi

2020

AAAI

View full text Add to dashboard Cite

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.

show abstract

Section: Evaluation Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

Zaltron

Zurlo

Risi

2020

AAAI

View full text Add to dashboard Cite

show abstract

“…It is created by an eye witness for whom the face is unfamiliar, and publicised by the police in the hope that it will be recognised by someone who knows the individual concerned. Unfortunately, composites are generally poorly recognised by people who are familiar with the person being portrayed (e.g., Davies & Valentine, ; Frowd, Carson, Ness, McQuiston‐Surrett, et al, ; Frowd, Carson, Ness, Richardson, et al, ; Frowd et al, ; Valentine & Davis, ; Zahradnikova, Duchovicova, & Schreiber, ). Most research on composite construction has consequently focused on identifying the reasons why these systems produce such poor likenesses, in order to improve their effectiveness as an aid for the police (e.g., Brown et al, ; Frowd et al, ; Ness, Hancock, Bowie, Bruce, & Pike, ).…”

Section: Introductionmentioning

confidence: 99%

Effect of racial bias on composite construction

Bhardwaj

Hole

2020

Applied Cognitive Psychology

View full text Add to dashboard Cite

Summary We investigated how prior bias about a face's racial characteristics can affect its encoding and resultant facial composite construction. In total, 61 participants (24 Europeans, 18 Indians living in India and 19 Indians living in Europe) saw a racially ambiguous unfamiliar face and were led to believe it was either European or Indian. They created a composite of this face, using EFIT6. Two groups of independent raters (one Indian, the other European) then assessed the apparent race of each composite. A different two groups (one Indian, one European) assessed each composite's degree of resemblance to the target face, to determine whether this was influenced by the constructors' initial categorisation of the target face as “own‐race” or “other‐race.” Composites appeared significantly more “Asian” or “European” according to the bias induced in their creators, but there was no evidence of any own‐race bias in the resemblance ratings for the composites.

show abstract

“…Initially those reports can include general descriptions of the suspect appearance, providing information about sex, race and approximate height and age. In some cases, an investigator works with the eyewitness to construct a facial composite of the offender, resulting in an image intended to resemble the facial characteristics of the perpetrator (McQuiston-Surrett, Topp, & Malpass, 2006;Zahradnikova, Duchovicova, & Schreiber, 2016). Those images are then sent to police officers working in the field, in order to narrow the investigation of suspects and assist on the identification of perpetrators.…”

Section: Introductionmentioning

confidence: 99%

“…Facial composites are created with the purpose of reducing the number of suspects and to help identify a perpetrator, relying on someone's ability to recognize his or her face (Zahradnikova et al, 2016). Accurate composites work well as an investigative tool, but unfortunately an extensive literature shows that eyewitness memory can also be unreliable (Lindsay, Mansour, Kalmet, Bertrand, & Melsom, 2011;Lindsay et al, 2013;Memon, Mastroberardino, & Fraser, 2008), and that composites generated by eyewitnesses are usually not easily recognized (Davies & Valentine, 2013;Wells & Hasel, 2007).…”

Section: Introductionmentioning

confidence: 99%

Eyewitnesses memory for faces in actual criminal cases: An archival analysis of positive facial composites

Saraiva

Castilho²,

Nogueira³

et al. 2017

Estud. psicol. (Natal)

View full text Add to dashboard Cite

Facial composites are crucial in the criminal justice system. In this archival study, we investigated the assumption that the success of facial composites depends partly on variables related to the crime, which either impairs or facilitates mnemonic processes. When a facial composite is successful in taking an offender to court it is sometimes archived as a positive facial composite, including a photo of the culprit and information about the crime. A total of 88 positive facial composites were investigated. The accuracy of facial composites was tested as a function of five variables related to the crime: type of crime, presence of weapon, retention interval, exposure duration, and disguise. Participants judged the resemblance of the perpetrators' photo with their correspondent facial composite. The results pointed out that only exposure duration was significantly associated with facial composites accuracy. Possible implications and future directions for research using archived facial composites are discussed.

show abstract

Facial composite systems: review

Cited by 11 publications

References 79 publications

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

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

Effect of racial bias on composite construction

Eyewitnesses memory for faces in actual criminal cases: An archival analysis of positive facial composites

Contact Info

Product

Resources

About