WO2007066316A1

WO2007066316A1 - Method for constructing a physiognomy

Info

Publication number: WO2007066316A1
Application number: PCT/IL2005/001302
Authority: WO
Inventors: Gil Gibli
Original assignee: Globes Publisher News (1983) Ltd.
Priority date: 2005-12-05
Filing date: 2005-12-05
Publication date: 2007-06-14
Also published as: EP1958160A1

Abstract

Method for sketching a face of a person, based on its recollection by an eyewitness, in which a set of two dimensional or spatial geometric forms are determined, such that each form or a combination thereof, corresponds to a specific part of a human face. Then a set of proportions between pre-selected geometric forms and a set of distances between pre-selected geometric forms are determined. The eyewitness selects a general, basic, facial outline that essentially matches the facial outline he remembers and then, a group of geometrical forms, selected by the eyewitness, is added to the outline, while, for each geometrical form, the eyewitness optimally matches the location, shape, size and orientation of the geometrical form to facial portions of the person, using the proportions, the distances and relative angles derived from the orientation of the geometric forms.

Description

METHOD FOR CONSTRUCTING A PHYSIOGNOMY

Field of the Invention

The present invention relates to the field of drawing identifiable objects. More particularly, the present invention relates to a method for sketching a face of a person based on its recollection by another person.

Background of the Invention

It is commonly agreed that the human's sight sense is the primary sense. One aspect of this notion is that humans have, in general, excellent facial recognition ability, which includes, among other things, determination of the age, sex, intention, mood, etc., from the perception of a face. In addition, humans can normally recognize and discriminate between many faces, while recognizing large numbers of unfamiliar faces after only a short exposure thereof. The aforesaid capability is utilized primarily by law enforcing forces, which put a lot of efforts in attempting to draw, or to compose, the face of a criminal suspect, based on its recollection by an eyewitness. The resulting drawings, sketches or photocopies, are, then, distributed to the field police personnel to assist them in capturing the suspect.

However, humans often have great difficulty recalling facial characteristics in sufficient detail to allow the generation of an accurate composite of the individual's face. Therefore, conventional composite- drawing creation procedures, which depend heavily on recall of specific features, are not yet always adequate.

In general, there are two basic ways to compose, or to sketch, a recollected face. The first way is using noticeable lines that relates directly to the five, "classical", elements (i.e., hair, eyes, nose, mouth and chin). The second way is using unlined photographs. It seems that persons recognize unlined photographs more easily, comparing to lined sketches, because the presence of lines in the prior art appears to impair memory. A notion of the two aforesaid ways is given hereinafter by reference to prior art.

In the case of sketch artists, several drawings have to be drawn, which relate to specific "wanted face", until the eyewitness is satisfied with the resulting face. Another technique, which is known as the "PhotoFit" technique, involves usage of 600 interchangeable photographs of facial parts, for picturing five basic features: forehead and hair, eyes and eyebrows, mouth and lips, nose, and chin and cheeks. Alternatives to PhotoFit have also been developed, which include, for example, the Multiple Image-Maker and Identification Compositor (MIMIC), which uses film strip projections; and the Identikit, which uses plastic overlays of drawn features to produce a composite resembling a sketch. A computerized version of the Identikit process (available from the Visitex Corporation. US 4,676,753 discloses an overlay based identification kit similar to Identikit. US 4,034,485 discloses a composite face apparatus and method. According to US 4,034,485, a plurality of stacks of opaque pages are utilized such that each stack of page bears a representation of a facial part (e.g., eyes), while each page thereof bears a unique representation of this part (i.e., particular shape of eyes). The stacks of pages are arranged along a binding to represent, in order, representations of hair, eyes, nose and mouth. An eyewitness is intended to flip the stacks of pages until obtaining a face that resembles to the face of the suspect. Likewise, there is also provided a plurality of overlays for long hair, mustaches etc. for use with the composite face. However, this method has the drawback of using fixed versions of eyes, hair, noses etc., such that only a general perception of the - 3 -

("wanted") face is obtained. Another drawback is that the eyewitness has to flip many pages, which may cause him to be confused, and sometimes even tired, as to the 'correct' face composition. US 4,045,883 discloses an identification kit and method. According to US 4,045,883, two pluralities of photographic facial portions are provided, each portion bearing one or more facial characteristics. The first plurality of photographic facial portions is related to full front face views, whereas the second plurality of photographic facial portions is related to profile face views that correspond to the full front face views. The portions in the two pluralities of photographic facial portions are in the same geometric scale. However, due to the use of pluralities of portions, US 4,045,883 suffers from the same drawbacks as US 4,034,485. In most cases, eyewitnesses do not remember small details of the major five elements of the face (hair, eyes, nose, ears and chin), such as the exact contour lines thereof, or the size of each element. On the other hand, eyewitnesses do tend to remember the general shape of the face, the relative size and location of the elements in the general shape of the face.

Because all of the prior art methods for composing a face of a person involve recollection of specific details of the commonly used five elements of a face, asking an eyewitness, again and again, to remember the exact contour lines of the five elements (while the witness usually remembers them only vaguely), is a common phenomenon.

The problem lays in that the eyewitness usually has, as noted before, a

'faded' memory of the face, which he tries to express in words to guide, e.g., a sketch artist, while the "wanted" face is being sketched, or composed. This kind of communication, between the eyewitness and the artist, or 001302

- 4 - composer, is a source for disinformation, confusion and that hardly can be bridged. Therefore, drawing a sketch that is based primarily on the intrinsic details of the major face elements might render the composing of the recollected face very difficult, and, in some cases, even impossible, and, therefore, in many cases, the resulting sketch is very inaccurate.

Another problem that characterizes the prior art is that the resulting sketch of the face, or the composed face, is only a raw sketch/composite that cannot reveal to its analyzer other details than those shown directly by the drawing, nor can it imply additional, 'hidden', or concealed, characteristics of the sketched/composite face.

Another problem that characterizes the prior art is that because a sketched, or a composite, face is based primarily on the recollection of small details of the five elements of the face, the prior art do not allow associating the face to a distinct section of the population or to some stereotype. Associating a face to a particular stereotype can be beneficial because such association can add additional 'dimension' to the "wanted" face, which can assist in comprehending the suspect.

It is therefore an object of the present invention to provide a method for sketching a face of a person, based on its recollection by another person, that minimizes the need for verbal description of the person's face. It is another object of the invention to provide a method for sketching a face of a person, based on its recollection by another person, that minimizes the effect of emotional state of the eyewitness on the sketched face. It is still another object of the invention to provide a method for sketching a face of a person, based on its recollection by another person, which allows obtaining an accurate sketch of the face without concentrating first on specific details.

It is yet another object of the invention to provide a method for sketching a face of a person, based on its recollection by another person, which assists in the comprehension of the related suspect. Other objects and advantages of the invention will become apparent as the description proceeds.

Summary of the Invention

The present invention provides a method for sketching a face of a person based on its recollection by another person.

The method of the invention is characterized by that the recollection of the head and the jutting parts thereof (e.g., forehead, nose, lower jaw, ears, etc.) by an eyewitness is translated into a sketched face by use of several basic geometrical forms that allow, e.g., a sketch artist, to focus on the general outline/shape of the head and, thereafter, to focus on the size and location of each facial part relative to the other facial parts. By using these basic forms, the method of the invention utilizes more efficiently the visual memory of humans. In addition, the inventor of the current invention utilizes what may be generally called, pseudo art rules, in particular those relating to the Cubism style of art, for organizing the various geometrical forms. 'Cubism' is a style of art that stresses abstract structure at the expense of other pictorial elements especially by displaying several aspects of the same object simultaneously and by fragmenting the form of depicted objects The method of the invention is further characterized by that it allows associating a person, whose face is sketched, to a particular group, or stereotype, of population. An exemplary group is a group of humans having round shaped faces.

The method of the invention is further characterized by that it allows enhancing the recollected face by providing a depth perception to the recollected face. The aforesaid enhancement allows, e.g., to reveal in the face 'hidden', or 'concealed' features, recesses and jutting areas other than the five major elements (i.e., forehead, eyes, ears, nose and lips). For example, the eyes may be recessed back, and a jutting area may be projecting cheekbones or a lower jaw. More specifically, the aforesaid enhancement gives clues as to facial features that cannot be derived using conventional methods.

The method of the invention is further characterized by that the sketch artist starts to sketch a recollected face by focusing on the general shapes, relative sizes and locations of major facial features, and then, as the questioning of the eyewitness progresses, the sketch artist gradually focuses on nuances in the face.

According to the present invention, a new 'graphical language' is used, for providing a communication means that allows a simple and easy communication between the sketch artist and the eyewitness. The graphical language is based on several geometric forms, such as, but not limited to, a circle, oval, triangle, rectangle, "A", "U", "V", and "T", and it utilizes the relatively good memory of humans for general shapes and for relative size and location of figural elements that are contained within a general, larger, figural element. The basic figures, which are (during the 005/001302

- 7 - questioning of the eyewitness) modified by the sketch artist as required to comply with the witness's recollection of the "wanted" face, allow the witness to easily and graphically express his general comprehension of the head/face of the suspect. By using these geometrical forms, the witness is not required to remember specific details of, e.g., the nose, eyes, etc. Instead, the witness is requested to associatively relate the corresponding geometrical figures to the general shape of the head and to the relative size and location of the five major facial elements contained therein, as recollected by him.

Despite of the fact that specific details of the facial elements have only limited weight in the sketching of the recollected face, it has been found by the inventor of this invention that eyewitnesses tend to remember such details while the sketch is in progress, because they subconsciously associate the lines drawn in the 'raw' sketch to these details. For example, it is most likely that an eyewitness will easily recall the location of the nose relative to the eyes, but not the exact details (e.g., exact shape or size) thereof. However, as the sketch artist progresses in his work, it usually occurs that, after the witness sees the sketch, the lines in the sketch remind him, in associative manner, at least some of the specific details.

The method for sketching by a sketch artist a face of a person based on its recollection by an eyewitness, comprises use of geometric forms as fiducial marks, the proportions, relative size and location of which essentially represent proportions, relative size and location of portions, parts, lines, and/or elements in the face recollected by the eyewitness.

More specifically, the method for sketching a face of a person based on its recollection by an eyewitness comprises: (a) determining a set of two dimensional, and/or spatial geometric forms, each of which, or a combination thereof, corresponds to a specific part of a human face; (b) IL2005/001302

- 8 - determining a set of proportions between pre-selected geometric forms; (c) determining a set of distances between pre-selected geometric forms; (d) allowing the eyewitness to select a general, basic, facial outline that essentially matches the facial outline he remembers; and (e) adding a group of geometrical forms, selected by said eyewitness, to said outline, while, for each geometrical form, allowing said eyewitness to optimally match the location, shape, size and orientation of said geometrical form to facial portions of said person, using said proportions, said distances and relative angles derived from the orientation of said geometric forms.

The forms are preferably the letters "A"- for characterizing the nose and its vicinity; "V"— for characterizing the Zygomatic Bones and the lower jaw; "U"— for characterizing the chin, rectangles, for characterizing the frontal bone, temples and nose; a set of general, basic, facial outlines, for characterizing the general shape of the recollected head; and 'bicycle seat'- like form, for further characterizing the lower jaw and said Zygomatic Bones.

It is noted that the 'bicycle seat'- like form is primarily intended for characterizing the structure, or bones, of the lower jaw, and, depending on the face of the person, the TDicycle seat'- like form might give also clues as to the characteristics of the Zygomatic Bones.

In a preferred embodiment of the present invention, the method for sketching a recollected head/face comprises the steps:

1) Relating the general shape of the recollected head to one of several general, basic, facial outlines, and modifying the proportions of the related basic facial outline as required to conform to the proportions of the recollected head. The one of several basic facial outlines that is IL2005/001302

- 9 - found by the eyewitness as most resembling the outline of the "wanted" head is hereinafter referred to as the 'sketched head'.

Preferably, the basic facial outline, for characterizing the general outline of the recollected head is selected from the group of: (1) Oval

Shape (OS), (2) Round Shape (RS), and (3) Triangular Shape (TS).

Since the OS type head is the most common figure, it must be further analyzed with respect to its different parts, or elements, such as cheek bones and jaw, an explanation of the characterization thereof is explained hereinafter. The 'RS' type of heads implies high probability of a 'full^* face, with tendency to complete a semi-circle that passes below the chin. The 'TS' type heads indicates conspicuous cheek bones and lower jaw, but there is a risk that if a specific head is covered with dens hair, the witness might erroneously remember it as belonging to the OS type.

2) Utilizing a drawing of a rectangular cup with holding 'ears' and proportions similar to the proportions of the sketched head, for relating the recollected ears to the sketched head and locating them on the sides of the sketched head while considering the general shape of the ears' auricle and their size and location relative to the sketched head. After the witness "impersonates" the 'ears' of the cup to the recollected ears, the witness associatively understands that he is requested to refer to the shape of the ears, and to their size and location relative to the vessel. Then, the witness can guide the sketch artist as to the location of the ears and their size relative to the sketched head. When the witness revives his memory in connection with the shape, relative size and location of the ears, additional details of the ears might 'popup' from his subconscious, such as details pertaining to the ears' lobule or helix. At this stage, the L2005/001302

- 10 - witness is fully aware of the figural relationship between the image of the object (i.e., the cup) and the recollected face;

3) Adding to the sketched head (with the recollected ears sketched as well) a rectangle, to symbolize the recollected neck. At this stage, the sketched head and the ears are shown to the witness overlaid on a rectangular figure that is associated with a basic neck. The association of the rectangle to a general neck is intended to show to the sketch artist the general shape of the recollected neck, and its location and proportions relative to the sketched head. This way, the witness can easily instruct the sketch artist to draw a neck with proportions recollected by him. For example, the witness may guide the sketch artist to change the width and/or height of the rectangle until the recollected proportions thereof are obtained. In case where the person involved is a woman, the two, substantially, vertical lines of the rectangle will be a little curved. In the latter case, the rectangle will look more like a trapezoid, the wider base of which faces downwards; i.e., to the direction opposite to the direction of the head;

4) Drawing and presenting to the eyewitness a main rectangular shape whose proportions essentially resemble the proportions of the sketched head, for symbolizing it, and, thereafter, drawing a line by the sketch artist for providing the main rectangle with a depth perception and adding three secondary, essentially parallel, rectangles, being an upper, middle and lower rectangles, which are circumscribed by said main rectangle. The upper drawer symbolizes the frontal bone and its lower portion symbolizes the area connecting the upper portion of the eyes' Supraorbit borders to the nasal bone (i.e., the nose 'bridge'). The left hand side and the right hand side of the upper drawer symbolize the Temporal Lines. The middle drawer is divided into three, not necessarily identical, horizontal portions, the middle portion being connected to the upper drawer to form IL2005/001302

- 11 - therewith a "T"-like part. The left and right hand sides of the middle drawer symbolize Zygomatic Bones. The secondary rectangles symbolize three 'drawers', each of which can jut independently of the other drawers, for symbolizing jutting parts and recesses in the recollected face. The three secondary rectangles are not necessarily identical to one another in their dimensions, and their prime purpose is to represent the proportions (i.e., as recollected by the eyewitness) of various facial parts that relate to the sketched head. For example, a jutting lower 'drawer' symbolizes a jutting lower jaw;

5) Obtaining the inclination angle (a) of the nose, relative to the Frontal Bone, by presenting to the eyewitness a side view of the main and secondary rectangles, and by causing the eyewitness to guide the sketch artist to jut one or more of the secondary drawers and 'turn' the middle drawer about an imaginary pivot that essentially coincides with a middle point in the area between the upper and the middle drawers, whereby to reconstruct the inclination angle (a) of the nose;

6) Drawing the letter "A" on the sketched head and utilizing it for characterizing the areas around the nose. The intersection point of the

"A" essentially coincides with a point in the nose bridge, whereas its two 'legs' symbolize the corresponding dimples between the cheeks to the sides of the nose. The 'baseline' of the "A" essentially coincides with the upper lip. The angle between the legs (β) provides meaningful information concerning the width of the mouth and the shape of the lips. The area that is delimited by the baseline of the "A", two legs of the "A" and the upper lip, provides information concerning the skin covering the upper teeth. The legs of the "A" figure provides meaningful information concerning the existence of distinct dimples; L2005/001302

- 12 -

7) Drawing the letter "U" to symbolize the chin as recollected by the eyewitness. Depending on the recollected chin, the "U" figure may be relatively wide, narrow, pointed or not;

8) Drawing the letter "V" such that each one of its two legs' starts at a point located between the left, or right, hand side eye, to the left, or right, hand side ear, respectively, and the two legs intersect in a point that can lay outside the sketched head, or on the "U" figure that symbolizes the chin. The "V" figure provides meaningful information concerning the two symmetrical opposite panels of the head, which include the cheeks, bones of the lower jaw and Zygomatic Bones;

9) Drawing a 'bicycle seat'- like figure, that includes a flat, symmetrically- shaped saddle and a seat post projecting downwardly from its middle portion. The seat is sketched to have two, horizontally, opposing ends and a middle end that projects downwardly towards the seat post. The seat post symbolizes the neck, the two opposite ends symbolize the Zygomatic Bones, and the middle end symbolizes the bone of the lower jaw. Accordingly, the proportions of the bicycle seat provide meaningful information concerning the structure of the cheekbones and the lower jaw, and also the relationships therebetween; and

10) Utilizing the preceding steps to further characterize the contour lines of the eyes' Supraorbital Border lines, the existence of dimples other than the dimples on the sides of the nose, the shape of the mouth and nose, age of the involved person (e.g., suspect), whether the suspect has a double chin, a short/long/narrow/wide neck, fleshy, or skinny, face, etc.

Wherein,

In each of the above-described steps, referring, by the sketch artist, to nuances in the description given by the eyewitness with respect to the recollected face, to obtain a more detailed image of the sketched head. Of course, the proportions, exact shape, size and location of the letters "A", "U", and the "V" figures depend largely on, and will be modified according to, the basic facial outline which the eyewitness considers as most resembling the facial outline of the head of the person.

Brief Description of the Drawings

In the drawings:

- Figs. IA, IB and 1C schematically illustrate three basic, or common, shapes of heads, according to the present invention;

- Fig. 2 schematically exemplifies two different ratios between height and width of a head;

- Figs. 3A and 3B schematically illustrate exemplary objects for demonstrating steps 1 and 2 of the method according to the present invention;

- Figs. 4A and 4B schematically illustrate two exemplary combinations of necks and heads, according to the present invention;

- Figs. 5A, 5B, 5 C and 5D schematically illustrate a cupboard and three drawers for teaching an eyewitness the figural language of the present invention;

- Fig. 6 schematically illustrates usage of the cupboard and drawers of Fig. 5 for obtaining features that relate to the forehead and nose;

- Fig. 7 schematically illustrates usage of the cupboard and drawers of Fig. 5 for obtaining the angle of the nose;

- Figs. 8A, 8B and 8C schematically illustrate three typical profiles of heads, for demonstrating three exemplary angles of noses as might be obtained using the drawing shown in Fig. 7;

- Fig. 9 schematically illustrates usage of the figural element "A" in accordance with the principles of the present invention; T/IL2005/001302

- 14 -

- Fig. 10 illustrates an exemplary sketched face, in which the figural elements "U" and "A" are joined together, in accordance with the principles of the present invention;

- Fig. 11 illustrates an exemplary sketched face in which the figural element "V" is exemplified, in accordance with the principles of the present invention;

- Fig. 12 schematically illustrates thick cheekbones;

- Figs. 13A and 13B schematically illustrate exemplary figural elements that are shapes as "bicycle seat", according to the present invention;

- Fig. 14 schematically illustrates another exemplary sketched face in which a figural element "V" is exemplified, in accordance with the principles of the present invention;

- Fig. 15 schematically illustrates determination of the eyes' supraorbit borders relative to the "A", in accordance with the principles of the present invention;

- Fig. 16 is a picture of the face of an exemplary person;

- Fig. 17 is a profile picture of the suspect whose face is shown in Fig. 16;

- Fig. 18 shows a cup and its two holding 'ears' that were sketched with reference to the face shown in Figs. 16 and 17;

- Fig. 19 shows a cupboard and upper drawer that were sketched with reference to the face shown in Figs. 16 and 17;

- Fig. 20 shows a preliminary head, chin and neck cupboard sketched with reference to the face shown in Figs. 16 and 17;

- Fig. 21 shows the lips as sketched with reference to the face shown in Figs. 16 and 17; and

- Fig. 22 shows the resulting profile sketch of the person shown in Figs.

16 and 17. 005/001302

- 15 -

Detailed Description of Preferred Embodiments

The set of two dimensional and spatial geometric forms comprises, for example, a sketched cup with 'ears' (Fig. 3B), rectangle 41 (Fig. 4A, shown modified in Fig. 4B), cupboard without and with drawers (Figs. 5A to 5D), the letters "T" (Fig. 6), "A" (Fig. 9), "U" (Fig. 10) and "V" (Fig. 14), 'bicycle seat' - like element (Fig. 13).

Figs. IA, IB and 1C schematically illustrate three exemplary general, basic, facial outlines: Oval Shape (OS) (Fig. IA), (2) Round Shape (RS) (Fig. IB), and (3) Triangular Shape (TS) (Fig. 1C), which can be utilized in accordance with the principles of this invention. Of course, other kinds of facial outlines can be utilized as well to characterize the general shape of the person's face/head. In step 1 of the method, the eyewitness is requested by the sketch artist to refer to the general shape of the head as recollected by him. In this step, the eyewitness chooses one of the aforesaid three basic outlines, or shapes, of heads; i.e., OS, RS or TS, as the shape that most characterizes the head. Thereafter, the eyewitness is requested to complete step 1 by modifying the proportions of the chosen outline of head, to conform to the recollected head. For example, if the eyewitness has chosen, e.g., the OS type (oval shape, Fig. IA) as the shape that most characterizes the recollected head (Fig. 2), he will be asked to complete this step by specifying to the sketch artist the recollected ratio between the height (H) to the width (W) of the head.

Referring to Fig. 2, the eyewitness may generally remember the head to be very narrow, as shown in Fig. 2A, or less narrow, as shown in Fig. 2B. The basic head's shape as chosen by the eyewitness serves as a general IL2005/001302

- 16 - larger figure, within which the facial elements will be arranged and characterized as describe hereinafter.

In step 2, the size and location of the ears are determined relative to the general shape of the sketched head (i.e., the head chosen by the eyewitness). This step is performed by presenting to the eyewitness schematic drawing of a vessel, such as a cup (30, Fig. 3A) whose general shape resembles the shape of the sketched head and, thereafter, adding a holding 'ear' to the cup (31, Fig. 3B).

Usually, the eyewitness associatively connects the 'ears' of the cup to the ears of the sketched head, and, accordingly guides the sketch artist as to the size and location of the ears relative to the sketched head. In this step, the sketched head is given a third (i.e., depth) dimension. From now on, the witness will understand the association of figural elements to the recollected head and will continue to direct the sketch artist without focusing on the image itself of the face, which could, e.g., traumatize the eyewitness. At least some of the major characteristics of the Zygomatic Bones are derived from lines (142 and 143, Fig. 14) that connect the upper portion of the ears of the person to his chin and pass on both sides of his mouth

In step 3, a neck is added to the sketched head and ears. A rectangular shape (41) is added to the sketched head (40, Fig. 4A), such that the traverse line 42 emphasizes the shape of the chin and lower jaw (43). This step is utilized for obtaining two main things: (i) the width (W) of the neck relative to the sketched head (Fig. 4A), and (ii) if the recollected neck is characterized by having straight, essentially parallel, lines, as shown in Fig. 4A (44, 45), it is most probable that the sketched head (40) belongs to a woman. However, if the lines of the neck are curved outwardly, as shown in Fig. 4B (46, 47), it is assumed that the sketched head (40) belongs to a man. Now, after utilizing steps 1 to 3 to characterize the 'external' features of the head of the "wanted" person, further steps are performed for characterizing, or determining, the internal features thereof. In the next steps, a reference is made to a rectangular shape to allow the recollected face to be divided into three main traverse portions, after which the main portions will be treated on individual basis.

In step 4, the sketch artist shows the eyewitness a main rectangular shape (Fig. 5A), whose ratio, between one of its sides to its neighboring sides, essentially resembles to the ratio between the height (H) and the width (W) of the sketched head (see Fig. 2). Next, a line (52) is drawn by the sketch artist (Fig. 5B), for providing the figure of Fig. 5A with perception of depth. Finally, three rectangles (53, 54 and 55) are added to the figure of Fig. 5B, as shown in Fig. 5C. The three rectangles do not necessarily have the same dimensions, and their purpose is to symbolize the proportions (i.e., as recollected by the eyewitness) of various facial parts that relates to the sketched head. Additional purpose of rectangles 53 to 55 is to provide depth perception, which is obtainable because, usually, eyewitnesses associatively identify rectangles 53 to 55 as three 'drawers', which can be (i.e., one or more of the drawers), therefore, Opened', independently of one another, to make the drawer(s) protrude from plane 56 (Fig. 5B). For example, a protruding 'drawer' 55 means that the sketched head, for example head 40 (Fig. 4), has a jutting chin bone. The upper drawer (53) corresponds to the forehead of the sketched head, and its lower part 57 corresponds to the part that connects the upper part 1302

- 18 - of the supraorbital border of the eyes and the upper part of the nose nasal. The sides (58, 59) of the drawer 53 represent the temporal lines.

The middle drawer (54) is related to the nose. Drawer 54 is partitioned to three parts (Fig. 6) 60, 61 and 62. The middle part (61) forms the vertical portion of a "T" (63), which represents the 'base' of "T" 63. The upper portion of "T" 63 overlaps the upper drawer 53. The "T" (63) can be differentially 'opened' by 'opening' drawers 53 and 54. By 'differentially' is meant that each of the drawers 53 to 55 (Fig. 5) can be opened to a different extent, and the "T" (63) gives an indication, or a clue, as to the depth of the eyes' supraorbital borders or to their 'shadowing' by the super ciliary's arches, and to the frontal bone. Fig. 5D schematically shows a profile of the 'drawers' 53 to 55, where each one of them is shown protruding to a different extent. The two sides of the middle drawer (54) represent the zygomatic bones (i.e., cheekbones).

The purpose of step 4 is to set the relative protrusion/depth of the forehead, nose, eyes and chin.

At this stage, it is still not established where the nose ends or what the purpose of the lower portion (66) of the middle drawer (54) is. However, significant information is obtained hitherto, part of which relates to the ratio between the 'thickness' of the nose to the width of the nose's nasal.

Step 5 is directed towards finding the inclination angle (a, Fig. 7) of the nose 73 relative to the plane of the forehead 71. In order to implement this step, an imaginary pivot 72 is virtually located in the upper portion of base

61 of "T" 63 (Fig. 6), in the gap between the upper and the middle drawers

(53 and 54, respectively, Fig. 5). The latter gap represents/designates the recess between the Supraorbital Border and the upper side of the nose nasal. At this stage the sketch artist is interested in the angle a (Fig. 7) of 005/001302

- 19 - the nose and, therefore, the proportions relating to the drawers 53 to 55 (Fig. 5) have, at this stage, only a secondary meaning. Additional information that can be derived from a sketch such as the one shown in Fig. 7 is the ratio between the length of the forehead 71 to the length of the nose 73. In Fig. 7, an exemplary ratio of 5:7 is shown.

Referring to the examples shown in Figs. 8A, 8B and 8C, a\ is relatively large, and is greater than al (Fig. 8B) and α3 (Fig. 8C). Therefore, in Fig. 8A the skin covering the upper teeth and the upper lip is shown 'pulled' (81) upwards, towards the nose, unlike the skin and upper lip in Figs. 8B and 8C.

Next, the angle between the forehead 85 to the nose 82 and the curvature of the forehead are determined. It is noted that, in accordance with the principles of the current invention, the shape, size and location of each facial element is determined, during the interaction between the sketch artist and the eyewitness, while considering the relationships between each of the facial elements to the other facial elements. With respect to Fig. 8, it will be simple and easy for an eyewitness to describe a remarkable nose, such as a flattened nose or a 'hooked' nose.

In step 6, the sketch artist focuses on the area locked' within the "A" figure (Fig. 9), as well as on the implications thereof on various areas that reside internal, or external, to the "A" figure. In particular, the sketch artist focuses on the lower portion of the "A", which represents the area below the ala of the nose. Referring to Fig. 9, the vertex (91), which can overlap the imaginary pivot 72 (Fig. 7), is located above the nose, and its base line overlaps the upper lip 92. The two legs 93 and 94 of the "A" represent the dimples that limit the area of the cheeks. The area 95, which is delimited by lines 93, 94 and 96, represents the skin that covers the upper lip 92, and, unlike in the prior art, is utilized as a facial element that stands for itself. It is noted that the "A" figure provides meaningful information, concerning the existence of distinct dimples, the characteristics of which depend on the thickness of the cheeks and on the age of the related person. The angle β corresponds to the width of the mouth, and also to the shape of the lips. For example, a narrow "A" (i.e., "A" with small β) emphasizes the recess between the nose's narises and the upper lip, and forms lips with "classical" shape. By 'small β' is meant that the length of the base of the "A", which essentially overlaps lip 92, is much shorter than its sides 93 and 94. In addition, a narrow "A" implies a fleshy lower lip.

Still referring to Fig. 7, an eyewitness is expected to modify angle a according to his recollection of the head. Angle a has implications on the vicinity of the nose, including the mouth/lips. In general, a larger (a) implies a smaller 'opening' angle β of an "A" figure, which is described hereinafter in connection with Fig. 9, that relates to the frontal view of the nose. A detail description of the utilization of the "A" figure is given in connection with Fig. 9 For example, for a relatively large a , the skin that covers the upper teeth tends to shrink upwardly, towards the nose, and to 'pull' with it the upper lip. This way, the recess between the upper lip and the bottom portion of the nose will be made noticeable.

In step 7, the sketch artist focuses on the "V" that is usually formed by two lines, each of which connects one of the two zygomatic bones to the chin. Referring to Fig. 14, it shows an exemplary head on which the "V" figure is shown to consist of legs' 142 and 143. The two legs 142 and 143 intersect each other outside the headl44; i.e., at imaginary point 141. Each leg of the "V" figure has a proximal end that lays essentially between the 2

- 21 - corresponding eye and the upper portion of the corresponding ear, and a distal end that lays on the "U" figure of the chin. Referring to exemplary 'leg' 142, its proximal end 148 lays between eye 145 and the upper portion of ear 146, and its distal end 149 lays on the "U" 147. In some cases, the "V" shaped figure might be very clear, like in the example shown in Fig. 14. Some people have a distinct "V'-shaped figure, whereas other people may have pseudo "V'-shaped figure, or one that somewhat resembles "V'- shaped figure but not completely 'V'-shaped figure. This part relates to the lower jaw and its determination is essential for completing the description of the face along the longitudinal axis of the head. The 'V includes the lower lip and its vertex lays on the lowermost point of the chin. In some cases, the 'V shaped figure might imply a projecting lower lip and/or chin, or a recess between the lower lip and the chin. The purpose of the above-described steps is to organize the various facial elements in proportional manner, while taking into account the sketched head type. However, it may occur, that, at a certain point, a decision will be made by the sketch artist, eyewitness, or both sketch artist and eyewitness, that the sketched head is the incorrect type and that it has to be replaced by a different type of head. However, after completing the seven steps, the results obtained therefrom can be utilized for further characterization of the face. That is, to the sketch, which is obtained after completing the seven steps, are added by the sketch artist with the consent of the eyewitness, facial features that are only implied by the already obtained features.

In step 8, the sketch artist focuses on the two, symmetrical, panels (101 and 102, Fig. 10) of the head. Basically, four factors are examined: (i) the location of the cheekbones relative to the head and to the other parts thereof, (ii) the characterization of the two lines of the 'V (97, Fig. 9), (iii) 005/001302

- 22 - the eyes' supraorbit borders (114, Fig. 10), and (iv) the estimated age of the sought person.

Each one of the above-described four factors is obtained according to its probability and with reference to the other obtained facial features. For example, if the sketched head is the RS type (Fig. IB), it is most likely that the person has fleshy face, and, therefore, it is most likely that the: (i) "V" is more like "U" (103) and he has a double chin (107) (ii) cheeks will be fleshy and rounded, (iii) "A" (104) will be relatively wide open (i.e., β - see Fig. 9 - will be relatively large), and its side lines (105 and 106) will curve with the cheeks and meet with the upper ends of "U" 103.

Referring to Fig. 11, it shows a case typical to a sketched head of the OS type. One end of a 'marker' straight line 110 coincides with the point 111 of the helix of the auricle 112, whereas the opposite end of line 110 coincides with a point 113 on the nose's ala. The lines of the "V" (108 and 109) are going towards to the chin. If the "U" (115, Fig. 11) is 'moderate', lines 108 and 109 will end essentially in the middle of the "U" (i.e., points 116 and 117).

In order to determine the shape of the lower jaw and whether the OS and TS shaped heads are more like square shaped, a notion of which is demonstrated in Fig. 12, "bicycle seat"- like figural element is presented to the eyewitness and utilized as described hereinafter in connection with Figs. 12, 13 and 14. The bicycle seat 139 includes a saddle 137 and a seat post 138. Saddle 137 includes two opposing horizontal ends 131 and 132, and a middle end 136 that projects downwardly towards the seat post 138. The determination process of the shape of the lower jaw, and whether the OS and TS shaped heads are more like square shaped, involves presenting to the eyewitness a the "bicycle seat" 139 (Fig. 13), in which the opposing ends 131 and 132 of saddle 137 symbolizes protruding areas in the lower jaw, as exemplified in Fig. 12 (121 and 122), and the middle end 136 symbolizes the exemplary lower jaw 123 (Fig. 12). It is noted that if the eyewitness does not recognize the figural element of Fig. 13A as a bicycle seat, it is possible to draw the seat as having two telescopic elements (133 and 134), which can be fixed to one another by a screwable element 135. The eyewitness is expected to relate the bicycle seat to the lower jaw and chin, and to the zygomatic bones, including to the relative location thereof in the sketched head.

It is noted that the form of the bicycle seat as presented to the eyewitness (Fig. 13A) is rather abstract and is not easy to grasp. That is, an eyewitness may not immediately understand that a bicycle seat is presented to him and how it is related to the face of the person. However, it has been found by the inventor of this invention that eyewitnesses usually understand that a bicycle seat is presented to them in light of the other, preceding, steps of the method of the invention. In cases where an eyewitnesse will not understand that a bicycle seat is presented to him, additional elements of the seat can be added to the sketch, for example elements 134 and 135 (Fig. 13B).

The eyes' supraorbit borders (102, Fig. 10) are determined by referring to two contour lines, as described in connection with Fig. 15. The two "L-I" lines start essentially from the corresponding inner intersection points (151, 152) of the upper and lower eyelids. "L-2" are lines that start from the aforesaid inner intersection points (151, 152) and continue downwardly, essentially parallel to the lines of the "A". The extent, to which these contour lines will be defined and emphasized, depends on the formerly obtained facial features, and on whether the cheeks are fleshy. In fleshy faces, the legs' of the "A" will be emphasized as a fleshy wrinkle, and the line "L-I" will be essentially the place where the cheek starts to thickens. Likewise, the line "L-2" will go downwardly, parallel to the corresponding line of the "A". In skinny faces, L-I will be emphasized, and line L-2 will go downwardly to the angles of the mouth and merge with the "U" 153 (Fig. 15).

Finally, the factors of the age of the suspect and the extent to which he is fat are considered as well. That is, in cases where the eyewitness remebers that the suspect is older than, say, fifty, and/or he is very fat, it will be assumed that his skin has wrinkles and its flesh is loosen and tend to drop to some extent, and the sketch will be amended accordingly. Of course, assuming that the suspect is relatively old and/or very fat can be utilized as a starting point.

A realistic example will be given hereinafter, in connection with Figs. 16 to 22. That is, the face of a real person will be sketched using the method of this invention. Referring to Figs. 16 and 17, they show a front and side/profile view, respectively, of a real person. An eyewitness will probably remember that the head of the person belongs to the round shape type (RS, see Fig. IB). Therefore, according to step 1 of the method, the sketch artist draws a three-dimensional cup (181, Fig. 18), the shape of which essentially resembles the proportions of the head shown in Fig. 16 (i.e., the RS shape). In addition, it is likely that the eyewitness will remember the ears as pointing outwardly (161 and 162, Fig. 16). Then, according to step 2 of the method, the eyewitness will guide the sketch artist to draw the recollected ears 161 and 162 as shown by reference numerals 182 and 183, respectively (Fig. 18). With respect to the exemplary person shown in Fig. 16, it may occur that the eyewitness will not remember the ears because they are very close to the head and tend to merge with the contour line of the head.

In step 3, the sketch artist will 'add' a neck to the head. However, the eyewitness might say that he does not remember a neck, because, as shown in Figs. 16 and 17, the neck seems to be very short and the sides of the head seem to merge with it (see also Fig. 4A, for comparison).

In order to allow the sketch artist to learn about the forehead of the person 160, as well as about aspects relating to forehead 166, the sketch artist draws, in accordance with step 4 of the method, a cupboard 191 (Fig. 19) with three drawers similar to the three drawers shown in Fig. 5. Only one drawer (190), which is relevant to the forehead in question (166) is shown in Fig. 19.

Referring again to Fig. 16, the upper 'drawer' 163, which is analogous to drawer 190 in Fig. 19 and to 'drawer' 53 in Fig. 5C, will be remembered clearly by the eyewitness because of the receding hairline and because of the broad forehead (166, Fig. 16). In addition, the two sides 164 and 165 of the 'drawer' 163 are very distinct, the 'drawer' 163 projects outwardly away from the plane of the face (towards the observer), which causes the eyes to recess back, whereby to obtain what is commonly known as 'deep set eyes', and the eyewitness may say that the ends of the eyebrows are pointed. Accordingly, the conspicuous presence of the eyebrows and the essentially continuous area underneath the lower line 167 of 'drawer' 163 will be reflected in Fig. 19 by the drawer 190 being 'pulled out' with respect to the plane, or surface, 193.

The drawing of Fig. 19 allows the eyewitness to point out to the sketch artist where the hairline starts and to refer to the receding hairline above the temples. According to this example, and utilizing the emphasized depth perception that the 'pulled out' drawer 190 makes, the eyewitness will guide the sketch artist to draw a line such as line 192 that corresponds to the hairline 168. In other words, the line 192 allows the sketch artist to conclude that the hairline of exemplary person 160 starts after the curve of the forehead.

The next step (i.e., step 5) is intended to determine the inclination angle of the nose 172 (see also angle a, Fig. 7). However, with respect to the exemplary person 160, speculations as to the nose inclination angle are irrelevant because person 160 has a flat nose 172 that is easily remembered.

The next step (i.e., step 6) involves utilizing the letter "A", and it is intended to determine the characteristics of the nose 172 and of its vicinity (see also Fig. 9). The letter "A" is shown in Fig. 16 quite distinct. The left and right legs' of the "A" essentially reach the opposing two ends of the mouth 170 and 171. The eyewitness will probably remember 'fleshy' lips. Considering the fleshy lips and the implication of the absence of neck (in step 3), the sketch artist learns that the neck merges with the round shape of the head, and that the chin is characterized by being relatively wide, both features are schematically shown in Fig. 20 (see reference numerals 201 and 202, respectively). If the eyewitness will remember a wide mouth line, this will be reflected by the baseline 173 of the letter "A" being relatively wide; that is, the angle β (see also Fig. 9) will be relatively large. Still referring to step 6, the lower thick, fleshy, lip will evidence a corresponding thick upper lip, and, accordingly, the sketch artist can conclude therefrom that the portions under the nose 172, which are circumscribed by a triangle, jut out towards the observer, and that there is a recess between the upper lip and the nose's ala, which leads to the conclusion that the upper line 203 (Fig. 21) of the upper lip is "classically" shaped.

As described hereinbefore, in connection with Figs. 12, 13 and 14, in order to determine the shape of the lower jaw and whether the OS and TS shaped heads are more like square shaped, as demonstrated in Fig. 12, a "bicycle seat" like figural element is utilized. However, with respect to the exemplary person 160, the "bicycle seat" like figural element is irrelevant because of the basically rounded and fleshy head of the person 160. In a fleshy face, it is most probable that the bone of the lower jaw is covered by relatively thick flesh and that the person has, what is commonly called as, a double chin, in which case the cheekbones will not be distinct. Accordingly, it is most probable that the eyewitness will not remember a distinct "V" (see Fig. 14). It is noted that, if the eyewitness would had remembered a distinct "V", this would had implied that the bone of the lower jaw was distinct, in which case the basic shape of the head would not had been rounded (i.e., the ES type) in the first place. The above embodiments have been described by way of illustration only and it will be understood that the invention may be carried out with many variations, modifications and adaptations, without departing from its spirit or exceeding the scope of the claims.

Claims

1. Method for sketching a face of a person based on its recollection by an eyewitness, comprising: (a) determining a set of two dimensional, and/or spatial geometric forms, each of which, or a combination thereof, corresponds to a specific part of a human face; (b) determining a set of proportions between pre-selected geometric forms; (c) determining a set of distances between pre-selected geometric forms; (d) allowing the eyewitness to select a general, basic, facial outline that essentially matches the facial outline he remembers; and (e) adding a group of geometrical forms, selected by said eyewitness, to said outline, while, for each geometrical form, allowing said eyewitness to optimally match the location, shape, size and orientation of said geometrical form to facial portions of said person, using said proportions, said distances and relative angles derived from the orientation of said geometric forms.

2. Method according to claim 1, wherein the forms are the letters "A"- for characterizing the nose and its vicinity, "V"— for characterizing the Zygomatic Bones and the lower jaw, "U"— for characterizing the chin, rectangles, for characterizing the frontal bone, temples and nose, a set of general, basic, facial outlines, for characterizing the general shape of the recollected head, and 'bicycle seat'- like form, for further characterizing said Zygomatic Bones and said lower jaw.

3. Method according to claim 2, further comprising:

a) Relating the general shape of the head of said person, as recollected by said eyewitness, to one of the general, basic, facial outlines, and modifying the proportions of the related basic facial outline as required to conform to the proportions of the recollected head, said related basic facial outline being a 'sketched head'; b) Utilizing a drawing of a rectangular cup with 'ears' and proportions essentially identical to the proportions of said sketched head, and causing said eyewitness to relate the recollected ears to said sketched head, by locating them on the sides of said sketched head while considering the general shape of the ears' auricle and their size and location relative to said sketched head;

c) Adding to said sketched head, with the recollected ears already sketched, a rectangle to symbolize the recollected neck, said sketched head and said ears are presented to said eyewitness overlaid on said rectangular, and said eyewitness guiding the sketch artist as to the neck's location and its proportions relative to said sketched head;

d) Drawing and presenting to said eyewitness a main rectangular form whose proportions essentially resemble the proportions of said sketched head, for symbolizing it, and, thereafter, drawing a line by said sketch artist for providing said main rectangle with a depth perception, and adding three secondary, essentially parallel, rectangles, being an upper, middle and lower rectangles, which are circumscribed by said main rectangle, said upper drawer symbolizes the frontal bone and its lower portion symbolizes the area connecting the upper portion of the eyes' Supraorbit borders to the nasal bone, said left hand side and the right hand side of said upper drawer symbolize the Temporal Lines, said middle drawer being divided into three, not necessarily identical, horizontal portions, the middle portion being connected to said upper drawer to form therewith a "T"-like part, the left and right hand sides of said middle drawer symbolize Zygomatic Bones, said secondary rectangles symbolize three 'drawers', each of which can jut independently of the other drawers, for symbolizing jutting parts and recesses in the recollected face, said three secondary rectangles are not necessarily identical to one another in their dimensions, and having a prime purpose of representing the proportions, recollected by said eyewitness, of various facial parts that relate to said sketched head;

e) Obtaining the inclination angle (a) of the nose, relative to the

Frontal Bone, by presenting to said eyewitness a side view of said main and said secondary rectangles, and by causing said eyewitness to guide said sketch artist to jut one or more of the secondary drawers and 'turn' the middle drawer about an imaginary pivot that essentially coincides with a middle point in the area between said upper and said middle drawers, whereby to reconstruct the inclination angle (a) of said nose;

f) Drawing the letter "A" on said sketched head and utilizing it for characterizing the areas around said nose, the intersection point of said "A" essentially coinciding with a point in the nose bridge, whereas the two legs' of said "A" symbolize the corresponding dimples between the cheeks to the sides of said nose, the TDaseline' of said "A" essentially coincides with the upper lip, the angle between said legs (/?) providing meaningful information concerning the width of the mouth and the shape of the lips, the area delimited by the baseline of said "A", the two legs thereof and the upper lip, providing information concerning the skin covering the upper teeth, said legs of said "A" providing meaningful information concerning the existence of distinct dimples;

g) Drawing the letter "U" to symbolize the chin as recollected by said eyewitness, depending on the recollected chin, said "U" figure may be relatively wide, narrow, pointed or not pointed;

h) Drawing the letter "V" such that each one of its two legs' starts at a point located between the left, or right, hand side eye, to the left, or right, hand side ear, respectively, said two legs having an intersection point that can lay outside said sketched head, or on said "U", said "V" providing meaningful information concerning the two symmetrical opposite panels of said sketched head, which include the cheeks, bones of the lower jaw and Zygomatic Bones; and i) Drawing a 'bicycle seat'- like figure, that includes a flat, symmetrically-shaped saddle and a seat post projecting downwardly from its middle portion, said seat is sketched such as to have two, horizontally opposing, ends and a middle end that projects downwardly towards the seat post, said seat post symbolizing said neck, the two opposite ends symbolize said Zygomatic Bones, and the middle end symbolizing the bone of the lower jaw, the proportions of said bicycle seat providing meaningful information concerning the structure of the cheekbones and the lower jaw, and also the relationships therebetween.

4. Method according to claim 3, wherein one step or a combination of steps is/are utilized to identify nuances in the face of the person.

5. Method according to claim 4, wherein the nuances relate at least to the contour lines of the eyes' Supraorbital Border lines, the existence of dimples other than the dimples on the sides of the nose, the shape of the mouth and nose, age of the person whose face is sketched, whether said person has a double chin, a short/long/narrow/wide neck and fleshy, or skinny, face.

6. Method according to claim 3, wherein the proportions, exact shapes, sizes and relative location of the letters "A", "U", and the "V" are modified as required to conform to the basic shape of the head which the eyewitness considers as most resembling the shape of the head of the person.

7. Method according to claim 3, wherein the basic facial outline is selected from the group of: (1) Oval Shape (OS), (2) Eound Shape (RS), and (3) Triangular Shape (TS).