Is it me, or is faces at best a beta / research quality feature?

Here are the problems I have encountered so far:
I'm in the same boat - never really used faces in iPhoto, just trying it out in Aperture. From what I can see, it's pretty weak in its implementation. A few of the things I have found:
The face detection itself is rather hit and miss. I've got photos that have a plain face that are not detected, and have had the moon, wheels, blankets, a horses butt and other odd things detected as faces.
I've tried to add a missing face, and it will sometimes save it, but I've got some photos that I have added the same face region 5 times and it won't save.
When I've tried to assign names to faces in bulk, I'll get no more suggestions, but when I go photo by photo, it will ask if it is the person I was trying to bulk update.
When you are assigning names photo by photo, it will occasionally decide it knows who it is, but there doesn't seem to be any shortcut key, so you need to stop using the keyboard, grab the mouse, select yes/no, then go back to typing - it's a bit of a pain.
The recognition is ... spotty. For some reason it decided that the Time cover of Barac Obama was my 2 year old daughter. It also thought she was the moon. And the blanket. And the horses butt.
When you are in the faces section of the library and you double click on a face, you can't get back to where you were. You need to go and select a node in the Library pane (like faces), then go back to the person you were looking at.
Have I mentioned that it misses lots of faces. Lots and lots and lots of faces. I've got one specific project with ~2100 pictures, it missed faces in about half of them. There isn't much point in using the feature if it isn't going to work for half of the photos, and given the awkward interface to manually find faces (and spotty at best persistence of edits), it's really not something worth doing.
To me, it seems like an interesting idea poorly implemented. Is anyone having any better success with the feature? Any tips or tricks to get it to work better?

Let's all face it, Faces is a "gee whiz" technology feature that uses copious amounts of processing power to display a technology that not only does not work very well, but even if it did it is not required by any professional photographer.
It cannot, and probably never will, be able to find ALL instances of a person being in a photograph, they don't claim it can, only human eyes using keywords can.
Now as a professional, as an example of where I want to identify some faces, I keyword all images for which I have model releases for the subjects, so they can be easily found and referred to. If I relied on faces it would only be finding SOME of these images, AND STEALING CPU TIME TO DO IT...grrrrrr!!! This is utterly useless and counterproductive.
I would like to hear from any other working professional on here, someone who derives 100% of their income from photography, please tell me I am missing something here, and I will immediately question your sanity...ER..no, I mean listen carefully to your opinion.
While I am on the soapbox, Places is of no professional use to me nor is geotagging...never had a client ask me or care about this ever...maybe in the future, who knows...
and Flickr and Facebook, jeez...ain't making me a dime...and why would I want to use these things professionally. How about a button for PROFESSIONAL SERVICES, like Pictage export manager in 64-bit at the A3 release, or Photo Shelter???
NO, THEY were relegated to PLUG-INS...according to you, Flickr and Facebook belong on the front page of a professional app...c'mon guys.
ALL THESE THINGS SHOULD BE OPTIONAL PLUG-INS!!! THEREFORE INCAPABLE OF CLUTTERING MY INTERFACE OR STEALING ONE SECOND OF MY TIME OR MACHINE PERFORMANCE....sorry, I will stop now, throat is getting sore from screaming...
THEY NEEDED TO STAY IN IPHOTO, where parents too lazy to keyword can get the computer to find their kids for them, a feature Apple even touted in its TV ads. "Stacks" from 2009 found here: http://www.apple.com/getamac/ads/. Having geotagged pics of the car trip from my son's coolpix was fun, and he made a map in this slideshow...more fun. Show me the money, and I will allow it on my interface.
Brushes and curves and all the presets JUST ROCK!!! So does the new converter, my Leica D-LUX 4 works now YAAY!...and so much more great stuff...HUGE SMILE HERE...
To the core photographic image management members of the Aperture team, it frankly must be professionally insulting to have to deal with marketing and product management who insist on putting c.r.a.p. features, useless to a professional shooter, on a superb professional product simply to attract technology buffs.
Ask marketing to see that focus group study of working pros who demanded these things...oops, musta missed that meeting, lol.
On top of that these gadgets DON'T EVEN WORK WELL, and you guys as a team are actually facing heat from users for things that had no business being there in the first place.
Faces impacts LIBRARY CONVERSION if left on...PERFECT, just PERFECT, as if you don't have enough to do finding and squashing version x.0 bugs in the real features without having to field calls about things that should never have been in Aperture 3 in the first place...jeez!
To the core photographic image management members of the team, YOURS is the work I highly respect and applaud and it is THAT GREAT WORK that needed to be highlighted, center stage and alone in the ads and on the box, not some useless, goofy feature like Faces and Places. I am indignant at this being here...truth to tell, I bet you are simply FURIOUS.
Sincerely,
K.J. Doyle
gotta go get a throat lozenge here...or maybe a stiff drink...

Similar Messages

Exporting photos for UHDTV or Native 4K TV, what are the best settings ? (File: Quality File: Color Space, Image Sizing and resolution) Or in other words; How can I get the smallest files but keep good quality for display on new UHDTV

Exporting photos for UHDTV or Native 4K TV, what are the best settings ? (File: Quality File: Color Space, Image Sizing and resolution) Or in other words; How can I get the smallest files but keep good quality for display on new UHDTV

You're welcome, and thank you for the reply.
2) Yesterday I made the subclips with the In-Out Points and Command-U, the benefit is that I've seen the clip before naming it. Now I'm using markers, it's benefit is that I can write comment and (the later) clip name at once, the drawback is that I have to view to the next shot's beginning before knowing what the shot contains.
But now I found out that I can reconnect my clips independently to the format I converted the master clip to. I reconnected the media to the original AVI file and it worked, too! The more I work with, the more I'm sold on it... - although it doesn't seem to be able to read and use the date information within the DV AVI.
1) Ok, I tried something similar within FCE. Just worked, but the file size still remains. Which codec settings should I use? Is the export to DV in MOV with a quality of 75% acceptable for both file size and quality? Or would be encoding as H.264 with best quality an option for archiving, knowing that I have to convert it back to DV if I (maybe) wan't to use it for editing later? Or anything else?
Thank's in advance again,
André

Where is the best place for suggestiong requesting features?

Where is the best place for suggesting/requesting features?
If something like this already exists, please let me know.
I often wish when I select a word to right click and choose dictionary to to find out the meaning of the word. Currently I have to copy, go to a dictionary web page, paste and search. And I spent a lot of time for that. I'm not that technically oriented to suggest how this can be achieved but just as an idea, may be Mozilla can partner with dictionary web sites by making a plug ins or include a dictionary with Firefox itself

I am sure there are Extensions that do this.
DictionarySearch - https://addons.mozilla.org/en-US/firefox/addon/68/
Add-on search for Dictionary - https://addons.mozilla.org/en-US/firefox/search/?q=dictionary&cat=all&lver=any&pid=1&sort=&pp=20&lup=&advanced=
Feedback - http://hendrix.mozilla.org/

Reg: Provide the best possible Scenarios regarding Planning feature of Analysis office as given below

Hello Everyone,
I am able to save the data back to the target from analysis workbooks using the Planning feature of the tool.
Please, can anyone Provide the best possible Scenarios regarding Planning feature of Analysis office.
Any answer is appreciable
Regards,
Karteek

Hi
Thank you for your post.
The reason why your contacts didn't display is currently the Tamil language is not supported by the Windows 7.5 OS and we do not have any information if this language will be supported in the future or not. The current list of supported languages are:
Brazilian Portuguese, Chinese (simplified and traditional), Czech, Danish, Dutch, Finnish, Greek, Hungarian, Japanese, Korean, Norwegian (Bokmål), Polish, Portuguese, Russian, and Swedish.
In regards to being a smart phone, these come from a variety of manufactures using a variety operating systems each having there own unique way of functioning and features and as such it is the responsibility of the consumers to make an informed choice on what device would suit them based on specifications of the device which includes hard ware or software. There is no unified smartphone requirements as each use's different operating systems.
So when purchasing a new device it is advisable to do some research on what device would suit your needs and if need be to speak to your local authorised mobile phone retailers for there expert advice. If at the point of purchase you were mis-advised and told that the Windows OS did support the Tamil language then this is something you will need to go back to the point of purchase otherwise it is your responsibility to make sure you choose a device that fully suit your needs.
Regards
If you find this post helpful, a click upon the white star at bottom would always be appreciated.
If it also solves your problem, clicking ACCEPT AS SOLUTION below it will benefit other users!

LRCC Face recognition - best practices?

Ok so we are all new to the wonderful world of face recognition in LR. I'm trying to work out what would be the best practices for using this.
A little bit of background - I have a catalog of over 200,000 images. In addition to portrait and wedding clients, a significant part of my work is with models and another significant part of my work is theatre photography. I have be wanting some sort of face recognition to help with both for some time.
What are your namining conventions for people? - here's mine:
Ideally I would label people as "surname, firstname" so that I can keep members of a family together in "named people" display, but commas are not allowed in names. Also the professional name of many models doesn't fit that pattern eg "Strawberry Venom" or "Cute as Sin" are to models I have worked with.
I am trying to come up with a sensible naming convention at the moment it is "Surname/ Firstname" for clients, theatre folk and friends/family. Models are still a problem, at present I am thinking of "Surname/ Firstname (model name(s))" While I may not be able to remember the real names of models, I do usually know the names from model releases. This naming will still permit me to filter/find them in the keyword List panel by just entering the model name.
On final addition I am making to this this naming convention is the use of a hashtag suffix to the name: #F for friends and family, #C for clients, #T for theatre/actors and #M for models. This enables me to filter on just models, or just actors, or just friends and familiy. Where people fall into multiple categories I add multiple hashtags. So photos of me would be keyworded with "Butterfield/ Ian #F #T"
Unknown / unidentified people.
What I am not yet certain about is how to handle unknown / unidentified people. Unidentified people fall into a number of different categories.
People I don't know and I am never likely to know (Eg random strangers on the street, local tour guides on holiday, random people in the background etc)
This group is relatively easy to deal with - that is to simply delete the face recognition, End of story.
People I don't know the names of yet but I am likely to find out (Eg actors in a production for which I don't have a programme)
For these people I am making up a unique name using the format "date/ Context-Gendernn" Eg an unknown male actor at Stockport Garrick Theatre would be named as "20150313/ SGT-M01" Although this may appear a complex solution it has a number of advantages. If/when I do learn the name of the individual (Eg I photograph them in a different production) it is simply a case of renaming the people keyword. Creating a unique name and not simple assigning all unknowns to a bucket name will help the face recognition algorithms find this person without it being confused by have different faces assigned to the same name. I am also using the hashtage #U to make it easier to filter the unknown faces when I need to.
People I don't know the names of and there is only a slim possibility of meeting/photographing again (Eg guests as a client weeding)
It feels as though I out to just delete the face recognition and have done with it, and this is what I would do except for thing. Other than manually drawing face regions I have not yet found a way to get lightroom to rescan a folder for faces if you have previously deleted the face recognition. This means that deleting face regions from a large number of people is something that cannot be easily reversed. I might just leave these people in the "Unnamed People" category... at lease until such time as there is a way to rescan a folder or colectoin.
Summary
My practices are still evolving. But I hope these thoughts and idea will help others think through the issues and come up with solutions that work for their situation. I am interested in hearing how other people are using the face recognition system. Especially if anyone is aware of any 'best practices' that Adobe or anyone else has recommended.

Glad it helped.
Yes and no. You can still put the people keywords into hierarchies within the keyword list - you can arrange them just like any other keywords.so you just create a "smith family" keyword and store "john smith" under it. What you can't do is apply BOTH smith family and john smith the the same face.
My use of the hash tags came about because I initially had a top level keyword for models, one for clients, one for theatre peple and one for family and firends. Then discovered that some of the theatre folk were also clients (headshots) and what to do when a friend is also a client. So the hash tag system means a person can be both a friend, a model, an actor as well as being a client! (#T #C #M #F).

ADF Faces & BC: Best pratices for project layout

Season greetings my fellow JDevelopers!
Our software group has been working with ADF for around 5 years and through the years we have accumulated a good amount of knowledge working with JDeveloper and ADF. Much of our current application structure has been resurrected in the early days of JDeveloper 10 where there were more samples codes floating around then there were "best pratice" documentation. I understand this is a subjective topic and varies site to site, but I believe there is a set of common practices our group has started to identify as critical to streamlining a development process(reusable decorated ui components, modular common biz logic, team development with svn, continuous integration/build, etc..). One of our development goals is to minimize dependency between each engineer as everyone is responsible for both client and middle layer implementation without losing coding consistency. After speaking with a couple of the aces at the last openworld, I understand much of our anticipated architectural requirements are met with JDeveloper 11(with the introduction of templates, declarative components, bounded task flows, etc..) but due to time constraints on upcoming deliverables we are still about an year away before moving on with that new release. The following is a little bit about our group/application.
JDeveloper version: 10.1.3.4
Number of developers: 7
Developer responsibilties: Build both faces & bc code
We have two applications currently in our production environments.
1.A flavor of Steve Muench's dynamic jdbc credentials login module
2.Core ADF Faces & BC application
In our Core ADF Faces application, we have the following structure:
OurApplication
     -OurApplicationLib (Common framework files)
     -OurApplicationModel (BC project)
          -src/org/ourapp/module1
          -src/org/ourapp/module2
     -OurApplicationView (Faces project)
          public_html/ourapp/module1
          public_html/ourapp/module2
          src/org/ourapp/backing/module1
          src/org/ourapp/backing/module2
          src/org/ourapp/pageDefs/
Total Number of Application Modules: 15 (Including one RootApplicationModule which references module specific AMs)
Total Number View Objects: 171
Total Number of Entities: 58
Total Number of BC Files: 1734
Total Number of JSPs: 246
Total Number of pageDefs: 236
Total Number of navigation cases in faces-config.xml: 127
Total Number of application files: 4183
Total application size: 180megs
Are there any other ways to divide up this application? Ie: module specific projects with seperate faces-config files/databindings? If so, how can these files be "hooked" together? A couple of the aces has recommended that we should separate all the entity files into its own project which make sense. Also, we are looking into the maven builds which should remove those pesky model.jpr files that constantly gets “touched”. I would to love hear how other groups are organizing their application and anything else they would like to share as an ADF best pratice.
Cheers,
Wes

After discussions over the summer/autumn by members of the ADF Methodology Group I have published an ADF Coding Standards wiki page that people may find useful:
[http://wiki.oracle.com/page/ADF+Coding+Standards]
It's aimed at ADF 11g and is intended to be a living document - if you have comments or suggestions please post them to the ADF Methodology google group ( [http://groups.google.com/group/adf-methodology?hl=en] ).

Best Codec to preserve quality when exporting media

Hi, I film with Canon XL H1 onto which I have bolted an AJA KiPro Mini Recorder. Via the SDI output I record in 10bit, 4:2:2 and ca. 200 Mb/s using the ProRes 422 (HQ) codec. I managed to get the .mov files onto my PC using Macdrive software and a special SYBA CF card reader. I use Windows 7-64bit and CS5.5. After editing I like to export media in the best quality possible. Best preset provided by CS5.5 seems to be AVCIntra 100. Am I doing the right thing? Better codec might be DNxHD, but this one is not proided as a prest. Like to store on HDD for later transferring to another suitable medium for potential acceptance by European tv stations.

Thanks Ann,
I am probably a total idiot, but never installed a codec before. Does the installer of say Lagarith install the codec in the correct directory? If not, where must I put it? Then in PrPro how do I access this codec when the drop-down menu of presets limits me to what is there?
Regards
Helmuth H. Fischer
Tel: +27- 82 800 8968; Fax: +27- 866 899 755
e-mail: [email protected]

Best Ways to research a FI/CO topic ???

Hello!
I am relatively new in the SAP consulting world - I have been an Oracle Financials Functional Consultant till now. Just got my SAP FI/CO certification a couple of months back (after 4 months hands-on training) and am only now getting to use real-world SAP as a Consultant.
From time to time, I need to do some research on a FI/CO topic for real-life client scenarios. I have been logging on to SAP Service Marketplace and have looked at various Portals on the Home Page, including the Help portal. However, it seems that I am not looking at the right research materials or relevant portals / forum discussions.
Could you please share best ways to do research on FI/CO topics?
What are the best ways to get to the comprehensive documentation and discussion threads on a topic, like FX Revaluation (how it works, how to configure, how to use, common issues etc), Intercompany Accounting / Reconciliation Issues etc. In some cases, I may even need to look up some basic functionality like Document Splitting, Use of Assignment field etc.
Perhaps I am not looking at right documentation on the Help portal. Where specifically on Help portal should I go to for FI/CO topics on ECC 5.0 or 6.0?
Any and all tips are welcome!
Puneet Nayyar

Hi,
Subsequent to posting the original message earlier today, I noticed that there was a related question posted by Amit on Oct 16th titled "Notes (Online) for SAP FICO Starter" to which there have been some responses.
Some of the tips I am looking for are already posted in Amit's question thread. You need not re-post those links.
If you have any additional tips to my original message, I would appreciate if you could post those.
Many Thanks!
Puneet

Best Practices for highly dynamic features like Search

For a project I need to implement a "Search" Component which will most probably use lucene that is built into CQ5. Since most of the other content on the site is dynamic and cached on dispatcher my concern is regarding the load such a dynamic feature will create for the publish instance.
What are the best practices to minimize the load on publish instance for such a scenario?

One option is to have your search results display via AJAX rather than a full page request. That way most of the page is cached in dispatcher and only the AJAX request with the search results is dynamic.

Best export for dvd quality using final cut express hd?

hey guys
i have final cut express hd...what would be the best export settings to get nice highquality dvd?
i am going to put about 25, short 5 minute clips in the dvd..will it be able to fit and all be a nice quality?
any suggestions?
thankyou!

Yes, provided you use dual layer DVD-R and your burner supports this newer media.
Otherwise the most you can burn to DVD-R SL media is 120 mins. (but keep in mind menus, titles, photos, etc all consume space as well thereby reducing the available amount of run time / QT Playback.

Best settings for highest quality for exporting a .mov file to h.264

Hi,
I want to know what the best settings are for the highest quality for exporting a .mov file (720p50) to h.264. I want the same quality as the original.
I want to know the data rate etc.
regards,
Jeroen
Message was edited by: Woudgraaf-AV

There is no way to get the exact same quality when transcoding from an editing source to a highly compressed H.264 QT. However, you can get a real high quality by adjusting the data rate.
However, these things depend on the length of the original. I routinely encode H.264s with good results in the 3000-5000 data rate range. I also make sure that I use multiple pass encoding. Take a short snippet of your project (10 seconds or so), and try various bit rates until you have one you like that doesn't take forever.
Andy

DVD Disk Space Requirements - Best Performance vs. Quality

I put together an iDVD 8 project that has about 135 minutes of content. When I go to burn the DVD using the Best Performance setting, is says it takes about 7gb (and I need to use DL). When I switch to the Professional Quality setting the disk space requirements are about half (almost to the point I can use a single-layer DVD). It seems it should be the exact opposite where more detailed, less compressed images should take more disk space.
Can someone explain this apparent contradiction? Thanks.

Google and learn about MPEG2 compression.... this is the backbone to iDVD. In general you have three encoding options:
CBR: Constant Bit-Rate
Single Pass VBR (variable bit-rate)
Two-pass VBR
CBR keeps the bit rate constant and therefore is the fasted way of encoding.
VBR will vary the bit rate. Scenes with a lot of motion get a higher bit rate, static scenes a lower bit rate.
Two pass VBR will analyze the entire sequence on the first pass to determine where and how high or low to go with the bit rate and then on the second pass, do the actual encoding. The two pass method may determine that you can go with a much lower average bit rate thus saving precious disc space. This method also takes twice as long!
But, that is just the math. Some sequences look better via CBR while others look better with two-pass VBR. Unfortunately, there isn't really a way to determine this prior to the lengthy encoding process. The best thing one can do is keep content as short as possible. If I had 135 minutes of material and I was using iDVD, I would want to create 2 DVDs instead of one. But that's me.
Mike

Best Performance vs Professional Quality

Which provides best playback quality?
Assuming the project is A) 40 min and B) 70 minutes, which one should I use?
Thanks

For under 60 minutes I always use Best Performance.
Best Performance encodes your audio/video at a fixed bit-rate that is about as high as a DVD player can play; so for content under 60 minutes, High Quality and Professional Quality can't really do any better.
High Quality uses a fixed-bit rate encoding level that is based on the playing time of all your content; the longer the content the lower the encoded bit-rate and the lower the image quality. Think of it as being approximately equality to Best Performance encoding at 60 minutes and getting slowly worse, the longer your content.
Profesional Quality use a two-pass method to vary the encoded bit-rate based on the image content. Areas with low detail are encoded at a low bit-rate making it possible to encode areas of high detail at a higher bit-rate. If your entire movie is filled with high detail/rapid action content, the results will look a lot like High Quality. If your content is a mix, then the high detail areas wlil look better than High Quality encoding and the low detail areas will look worse (but you may not notice). Profesional Quality encoding is much slower because of the extra pass evaluating the data.
For 60 minutes, I'd use Best Performance. For 70 minutes, I'd use High Quality.

Struts-faces nightly build, RI Beta

Hi,
I'm using struts-faces and am trying (desperately) to upgrade to the Beta RI of JSF. I got last night's build but I'm getting this error:
2004-01-13 17:19:35 Exception sending context initialized event to listener inst
ance of class com.sun.faces.config.ConfigListener
java.lang.NullPointerException
at com.sun.faces.config.ConfigParser.parseConfig(ConfigParser.java:93)
any ideas?
thanks.

Oops.
Forgot that I needed jsf-impl.jar in my classpath now.
Thanks!

Getting the best possible output video quality in Premiere 8.0

Here is my PC information:
Windows 7 64 bit
Intel Quad Core Q6600 @2.40 Ghz
4.0 GB of Ram
My Camera information:
Kodak Playsport HD
Records videos in HD 720P @ 60 FPS
Saves movies in .MOV format, 1280 x 720 resolution, and H.264 codec
Now I am playing around with the settings but when I edit movies in Premiere Elements and then save it I always lose a lot of quality. I'm not really interested in sharing the videos on the web, I'm just looking for the best possible quality (as close to how it looks in its original form) for viewing on my home computer. Can anybody help me out with decent export settings that will give me the best overall quality?

Hello Bill and thank you for your respone, I appreciate it. I've managed to get the overall quality of the movie a lot better with using different presets at new project. I am now using the preset NTSC-HDV-HDV720p 30 and that seems to be working a lot better. I also save the final movies as a Quicktime file, with a Local Area Network (LAN) and then I go to advanced and bring up the quality to 100, change it to 1280x720, bring the frame rate up to 60, and set the bitrate at 8000.
I am now having a different problem however. It seems that when I add a music to the background the song messes up frequently. What I mean is that when I'm watching the movie I created the video will be playing normally and then at random times the song will skip or get choppy. Then on top of that, with certain transitions between scenes the transition won't even work, it will just freeze for a bit and then skip to the next scene (On a side note, I leave the "Field Type" as "Progressive" and the "Aspect" as "Square Pixels [1.0] simply because I'm not sure what those are for).
I just uploaded this practice video to youtube, if you want to check it out it's at http://www.youtube.com/watch?v=eVorCPASaGM
Keep in mind I just threw together a couple random pictures and clips just to practice haha. You will be able to notice that the song and video get slightly choppy for some reason.
Thanks in advance!

Is it me, or is faces at best a beta / research quality feature?

Similar Messages

Maybe you are looking for