How can we improve Mobile Photography?
Cameras, software and infrastructure?
Mobile Photography= cell phones/vehicle mounted/aerial/blimps
Focus more on the desired change in
technology/algorithms/infrastructure rather than applications based on
what is already available. We dont want to simply react to existing
technology. We want to influence the next generation of applications.
Imagine IMAGE-SEARCH is solved, imagine Installed-Camera infrastructure
is available (or suggest more solutions to improve
image-search/infrastructure)
Technology:
What component of technology is currently poor and what needs to be
improved? What parallel developments in sci-tech can be borrowed
(negative index of refraction/3D cameras/gps/ID/)? What will the
ultimating mobile imaging device look like (because it will be carried
by humans, the way they will hold it/use it)?
Better illumination? Fantasy optics (flat)? Multispectral/thermal
imaging, live feed (TV), distant/night vision, 10x-performance (a
microscope in every phone), watermarking, geo-tagging
Applications
Video blogging, Citizen Journalism (Will CNN go bust), Augmented
reality (overlay), home monitoring/maintenance (HVAC, structure),
‘leave-your-phone-behind-intentionally’, video-pedia (video wikipedia),
shopping tele-buddy, instant product feedback (how fresh is the fish I
am buying), Video-twitter, live mobile video conferencing
Is there a way classify the applications? (We will make an attempt, but
suggest your taxonomy)
Emphasize on how we can develop new devices/software/infrast to
fundamentally change these applications.
Society:
Social networks, Online dating, News, Commerce, Environment, Politics
(e-voting on issues), Personal relationships, Agriculture,
Positive/Negative influences (hope/hate/discrim) (trust between Kenyan
rioters)
What are the parameters on the axis of exploration (e.g. Eugene talked
about local/remote sync/async last week, or 'functional imaging' (how
fresh is the fish))?
= = = = = = = =
HW Discussion Summary
Technology
-image quality (on par with standard cameras)
-new lens models
-portability compromises
--improved i/o from small devices
--projected interfaces/vr glasses
-tamper-proofing
-expanding modalities (more than visual info)
--spectra/scales/info(biometric,ad)
--stereoscopic input/output
-blackout for privacy
-smart/auto mms / remote storage
-localized (rss) content return
-capture recently viewed event
-optical nerve graft (get human eye camera for free)
Functional Imaging - more functional
-Animal Imaging - new perspectives, unreachable (for humans) locations
-Content Aggregation
--news
--crowd sourced views
--video-pedia (image indexed)
--good/bad drivers
-image search/identification
--gaze based
--shopping
Social Imaging
-video diary/blog/twitter
-meeting/contacts/friends/dating
--statistics based
--photo-exchange
-public space evaluation
-social watchdog - good behavior
Class Discussion Summary
questions for next week
ps 2028
- image search solved
- etc
- any multimedia manipulation tool
text audio video what's next?
whatever the next multimedia is, capture/edit it
new multimedia - whole worlds, sl
model future babies - upload parameters, genome
tools that capture materials - know nano structure
-- printers that can create structure
-- any material organic/inorganic can be captured, printed
-- designer materials for housing, electronics, self assembling furniture
-- video streams from body captured coin sized petabyte disk
---- perhaps used by doctors
-- break down in terms of tech, social impact
-- new ui? minority report? magic wand?
-- collaborative imaging, optics, illumination
-- instant software update of the future
---- instantly get software for new events
-- will lighting get professional?
-- video blogging
-- citizen journalism
-- cloning and beautification of tangible objects
get everything in on Tuesday by 5pm
-- will give others time to post comments
-- cross discussion - comment on each others responses
discuss cool ways to view new ml construction
-- fusion of ir, collaborative ui?
topic for survey paper by mar. 19
-- ideas for class
-- new ideas or from class
-- project or survey paper
-- 40% tech, 30% future 30% applications
-- individual paper, project team possible?
-- reqs. will be posted
Brandon's presentation - Beyond the visible spectrum
-- Thermal IR - not night vision
-- Tom's question - IR imaging invariant to day/night?
---- many applications looking at delta in temp
-- Fredo - IR sensitive film closer to visible domain (near IR) - thermal IR is much lower freq.
---- demos of near IR photos
-- Fredo - Sony night vision is IR (night shot)
---- red wine/coke/some clothing transparent in near IR
-- UV imaging
-- Millimeter wave imaging
---- tsa.gov lots of hype, not much info
-- Summary
---- adaptive environmental lighting - estimate lighting
Nokia Talk (I'm assuming that the slides will be available)