One of the reasons face detection and recognition has attracted so much
research attention and sustained development over the past few decades is
its great potential in numerous government and commercial applications. In
1997, at least 25 face recognition systems from 13 companies were available
and since then, the numbers of face recognition systems and commercial
enterprises have greatly increased owing to the emergence of many new
application areas. Although some of these techniques are not publicly available
for proprietary reasons, one can say that many others have been reported in
research publications and available as affordable commercial systems.
The technology has evolved from laboratory research to many small,
medium or large commercial deployments. At present, it is most promising for
smaller medium-scale applications, such as office access control and computer
log in; it still faces great technical challenges for large-scale deployments such
as airport security and general surveillance.
Notwithstanding the extensive research effort that has gone into the area
of face detection and recognition, a system that can be deployed effectively
in an unconstrained setting is yet to evolve due to much variability in image
parameters. Moreover, sensors and image capturing techniques also play a
vital part in the overall success of a system. Another direction for improving
recognition accuracy lies in a combination of multiple biometrics and security
methods. It can work with other biometrics such as voice-based speaker
identification, fingerprint recognition and iris scan in many applications.
The only system that does seem to work well in the face of these
challenges is the human visual system. It makes sense, therefore, to attempt
to understand the strategies this biological system employs, as a first step
towards eventually translating them into machine-based algorithms. We
believe that the future efforts may likely use such a path for refinement and
performance improvement of present face recognition systems. Insights into
the functioning of the human visual system may serve primarily as potentially
fruitful starting points for computational investigations.
The issue of connectivity between the perception under the domain of
cognitive science and the computer vision technique under the domain of
computer technology is perhaps one of the interesting and current issue in face
recognition. An interesting example is given in the associated figure to explain
the statement. In this figure, the issue of perception will decide whether we are
seeing the face image of Shakespeare or a landscape image consisting of trees
304 Conclusion
FIGURE 14.6: Visual perception and recognition illustrated
and cottages. Nothing has been said on this issue in this book. The central
issue involved in this type of problem is related to mind-body interaction.
The question arises regarding the process involved by which the brain - the
material object of our body - can evoke correct signals in our mind, which in
turn may control many of our actions or inactions for detecting or rejecting a
face. This may inspire additional useful curiosity in intelligent face detection,
which may go beyond the hard scientific or technological issues to the areas
of reasoning and consciousness.
In the present book, the issues of realities of the mind-body problem
have not been touched upon, but the realities which are under the realm
of technology are discussed and investigated. An attempt has been made to
realize the methodologies that can detect and recognize faces, even with some
deviations in a limited sense of the practical world. Incidentally, using the
human system for such detection and recognition is one aspect of coordination
between visual perception, intelligence and muscle action which is related to
the mind-body problems. However, one should be conscious and accept that
the two kinds of realities are interdependent, which means that there can be a
correlation between the two. Perhaps a day is not far off when these questions
would be addressed in the language of science and translated in the domain
of technology. Perhaps at the end, it may be quoted from the aphoristic book
Tractatus Logico-Philosophius by L. Wittgenstein, ”what we cannot talk about
we must pass over in silence.

Get Face Detection and Recognition now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.