Tuesday, May 28, 2013

May Newsletter - Video Analytics


The Vega newsletter is published monthly by Vega BI, and distributed to our Brazilian partners to facilitate pursuit of a common interest in top-notch technologies. 
Video Analytics

Video content analytics (VCA) is the capability of automatically analyzing video to detect and determine temporal events not based on a single image. As such, it can be seen as the automated equivalent of the biological visual cortex.
This technical capability is used in a wide range of domains including entertainment, health-care, retail, automotive, transport, home automation, safety and security. The algorithms can be implemented as software on general purpose machines, or as hardware in specialized video processing units.
Many different functionalities can be implemented in VCA. Video Motion Detection is one of the simpler forms where motion is detected with regard to a fixed background scene. More advanced functionalities include video tracking and egomotion estimation.

Based on the internal representation that VCA generates in the machine, it is possible to build other functionalities, such as identification, behavior analysis or other forms of situation awareness.
VCA relies on good input video, so it is often combined with video enhancement technologies such as video denoising, image stabilization, unsharp masking and super-resolution.


For more .... Download Vega's May Newsletter

Sunday, May 19, 2013

Big Data analytics to make "House of Cards" a hit


Giving Viewers What They Want

By DAVID CARR, New York Times

In the television business, there is no such thing as a sure thing. You can have a gold-plated director, a bankable star and a popular concept and still, it’s just a roll of the dice.
Or is it?
In any business, the ability to see into the future is the killer app, and Netflix may be getting close with “House of Cards.” The series, directed by David Fincher, starring Kevin Spacey and based on a popular British series, is already the most streamed piece of content in the United States and 40 other countries, according to Netflix. The spooky part about that? Executives at the company knew it would be a hit before anyone shouted “action.”
Netflix, which has 27 million subscribers in the nation and 33 million worldwide, ran the numbers. It already knew that a healthy share had streamed the work of Mr. Fincher, the director of “The Social Network,” from beginning to end. And films featuring Mr. Spacey had always done well, as had the British version of “House of Cards.” With those three circles of interest, Netflix was able to find a Venn diagram intersection that suggested that buying the series would be a very good bet on original programming.
Big bets are now being informed by Big Data, and no one knows more about audiences than Netflix. A third of the downloads on the Internet during peak periods on any given day are devoted to streamed movies from the service, according to Sandvine, a networking provider. And last year, by some estimates, more people watched movies streamed online than on physical DVDs.
Film and television producers have always used data, holding previews for focus groups and logging the results, but as a technology company that distributes and now produces content, Netflix has mind-boggling access to consumer sentiment in real time.
How much data does it have at its fingertips? According to GigaOm, Netflix looks at 30 million “plays” a day, including when you pause, rewind and fast forward, four million ratings by Netflix subscribers, three million searches as well as the time of day when shows are watched and on what devices.
Jonathan Friedland, the company’s chief communications officer, said, “Because we have a direct relationship with consumers, we know what people like to watch and that helps us understand how big the interest is going to be for a given show. It gave us some confidence that we could find an audience for a show like ‘House of Cards.’ ”
In addition, movies and TV shows on the service are annotated with hundreds of tags — metadata descriptors — inserted by viewers commissioned to describe the talent, the action, the tone and the genre, among many, many other things. In the past, those tags were used to recommend other shows from the long tail of content on the service, essentially building profiles based on the preferences of individual subscribers. But now Netflix is commissioning original content because it knows what people want before they do. “There are 33 million different versions of Netflix,” said Joris Evers, the company’s director of global corporate communications.
Based on that information, Netflix bought “House of Cards.” It is also producing new episodes of “Arrested Development,” and in April, it will begin streaming episodes of “Hemlock Grove,” a horror-thriller based on a novel of the same name.
Netflix has always used data to decide which shows to license, and now that expertise is extended to the first-run. And there was not one trailer for “House of Cards,” there were many. Fans of Mr. Spacey saw trailers featuring him, women watching “Thelma and Louise” saw trailers featuring the show’s female characters and serious film buffs saw trailers that reflected Mr. Fincher’s touch.
It is impossible to say that “House of Cards” is a hit because Netflix, to the consternation of some of its more traditional competitors, is not participating in ratings. But social media is thick with mentions of both the new programming and the new paradigm. The show made the front page of The New York Times and The Los Angeles Times, and was on the cover of Emmy magazine, a good omen for its awards future. And when your price is as low as Netflix’s — $7.99 a month for streaming — a flurry of buzz can pull plenty of people off the fence.
While careers and entire networks have been made and lost based on the mysterious alchemy of finding a hit, Netflix seems to be making it look easy, or at least making it a product of logic and algorithms as opposed to tradition and instinct.
A cable executive who has talked to Amazon says that its Prime service, a nascent effort to get into original content, will also lean hard on data-driven approaches to determine its programming. The executive, who asked not to be identified because the discussions were private, said it would change the way that business operates sooner than people thought.
“I think it is a little hysterical to say that Big Data will win the day now and forever, but it is clear that having a very molecular understanding of user data is going to have a big impact on how things happen in television,” he said.
Others aren’t so sure. John Landgraf, who, as president and general manager of FX Networks, has had a good run at the channel in finding hits, said he thought numbers-crunching would never have predicted the success of “The Sopranos,” “South Park,” and “Mad Men,” among others, including hits he has said yes to, like “Sons of Anarchy.”
“Data can only tell you what people have liked before, not what they don’t know they are going to like in the future,” he said. “A good high-end programmer’s job is to find the white spaces in our collective psyche that aren’t filled by an existing television show,” adding, those choices were made “in a black box that data can never penetrate.”
The rise of the quants has some worried about the impact on quality and diversity of programming. Writing in Salon, Andrew Leonard wonders “how a reliance on Big Data might funnel craftsmanship in particular directions. What happens when directors approach the editing room armed with the knowledge that a certain subset of subscribers are opposed to jump cuts or get off on gruesome torture scenes” or are just interested in sexual romps?
Netflix insists that actual creative decisions will remain in the hands of the creators. “We don’t get super-involved on the creative side,” Mr. Evers said. “We hire the right people and give the freedom and budget to do good work.” That means that when Seth Rogen and Kristen Wiig are announced as special guests on coming episodes of “Arrested Development,” it is not because a statistical analysis told Netflix to do so.
But there are potential conflicts. Given that Netflix is in the business of recommending shows or movies, might its algorithms tilt in favor of the work it commissions as it goes deeper into original programming? It brings to mind how Google got crossed up when it began developing more products, and those began showing up in searches.
And there are concerns that the same thing that makes Netflix so valuable — it knows everything about us — could create problems if it is not careful with our data and our privacy. But many think the trade is worth it.
“Netflix and Amazon know when you stop and start a program, whether you wanted the whole thing, all of that,” said Rick Smolan, whose most recent book was “The Human Face of Big Data.” “Programmers have been wandering out and shooting a shotgun into the night sky and hoping they hit something, and I end up paying $150 for channels full of nothing I want to watch. These guys know what they are aiming at.”
Netflix’s command of data, including mine, isn’t foolproof. It thinks I like “The West Wing,” which I don’t, and it thinks I am a sucker for every quirky little indie movie that floats in, which I am not. But when it came to guessing if “House of Cards” might appeal to me — politics, media and Mr. Fincher are all hot buttons — the deck was stacked in its favor.
Not long after the series became available, I found myself in a dark room, surrounded by empty food wrappers and unmet deadlines, wondering when the second season was going to start. I never had a chance.

Friday, May 10, 2013

Video Analytics - In a Nutshell


 What is Video Analytics?

Video Analytics, also referred to as Video Content Analysis (VCA), is a generic term used to describe computerized processing and analysis of video streams. Computer analysis of video is currently implemented in a variety of fields and industries, however the term “Video Analytics” is typically associated with analysis of video streams captured by surveillance systems. Video Analytics applications can perform a variety of tasks ranging from real-time analysis of video for immediate detection of events of interest, to analysis of pre - recorded video for the purpose of extracting events and data from the recorded video (also known as forensic analysis). Relying on Video Analytics to automatically monitor cameras and alert for events of interest is in many cases much more effective than reliance on a human operator, which is a costly resource with limited alertness and attention. Various research studies and real-life incidents indicate that an average human operator of a surveillance system, tasked with observing video screens, cannot remain alert and attentive for more than 20 minutes. Moreover, the operator’s ability to monitor the video and effectively respond to events is significantly compromised as time goes by.
Furthermore, there is often a need to go through recorded video and extract specific video segments containing an event of interest. This need is growing as the use of video surveillance becomes more widespread and the quantity of recorded video increases. In some cases, time is of the essence, and such review must be undertaken in an efficient and rapid manner. Analyzing recorded video is a need that can rarely be answered effectively by human operators, due to the lengthy process of manually going through and observing the recorded video and the associated manpower cost for this task.
While the necessity for, and benefits of, surveillance systems are undisputed, and the accompanying financial investment in deploying such surveillance system is significant, the actual benefit derived from a surveillance system is limited when relying on human operators alone. In contrast, the benefit accrued from a surveillance system can be significantly increased when deploying Video Analytics.
Video Analytics is an ideal solution that meets the needs of surveillance system operators, security officers, and corporate managers, as they seek to make practical and effective use of their surveillance systems.

 What is Video Analytics used for?

Video surveillance systems are typically installed to record video footage of areas of interest within a facility, around its perimeter or in outdoor areas which require observation, with a view to “catching” (allowing the user to be able to observe) and recording events related to security, safety, loss prevention, operational efficiency and even business intelligence.
Video Analytics enhances video surveillance systems by performing the tasks of real -time event detection as well as post-event analysis, while saving manpower costs and increasing the effectiveness of the surveillance system operation.

 Video Analytics for Real-Time Alerts

Through the implementation of various image processing algorithms, Video Analytics can detect a variety of events, in real-time, such as:
  • Penetration of unauthorized people / vehicles into restricted areas
  • Tailgating of people / vehicles through secure checkpoints
  • Traffic obstacles
  • Unattended objects
  • Vehicles stopped in no-parking zones, highways or roads
  • Removal of assets
  • Crowding or grouping
  • Loitering

By defining the set of events that the surveillance system operator wants to be alerted to, the Video Analytics software continuously analyzes the video in real-time and provides an immediate alert upon detection of a relevant event.

 Video Analytics for Investigation

In addition, Video Analytics algorithms may be implemented to analyze recorded video, a task that is challenging and time consuming for a human operator, especially in cases whereby a large amount of video must be reviewed. Through rapid analysis of recorded video, Video Analytics can perform the following tasks:
  • Pinpoint an event in recorded video, and retrieve the relevant video segment from the stored video
  • Perform analysis of motion patterns and detection of motion irregularities in defined areas
  • Perform a variety of statistical analysis tasks relating to people or vehicles over defined periods of time

Through the use of search queries, the surveillance system operator defines the event or analysis desired in a specific segment of recorded video. The Video Analytics system analyzes the video and provides the search results through an automated search, without requiring any additional intervention from the operator.

How video analytics helps reconstruct Boston Marathon bombings

On Thursday morning, authorities were reportedly set to release photos of two suspects in the bombings, although the analysis of all that footage will undoubtedly continue, as the police and FBI seeks to piece together the chain of events. The FBI released photos of the suspects Thursday night, asking for help in identifying them. Later, one of the suspects, identified as brothers
How do investigators weed through terabytes of video in different formats, whether 30-second snippets from cell phones or hours of footage from a surveillance camera at a nearby store? Going through all that footage is still largely a labor-intensive task, but video analytics and digital forensics tools can help investigators compress video, pinpoint areas of interest, look for anomalies and find relevant details, according to government and industry experts.


Tuesday, May 7, 2013

BYOD (Bring Your Own Device) Needs or False ?



Where's the BYOD Payoff?
Companies may be bleeding corporate dollars in the name of BYOD productivity gains that don't really exist, says Nucleus Research.
 By Tom Kaneshige
Wed, April 24, 2013

CIO — Companies jumping on the bumpy "Bring Your Own Device" bandwagon might be the real losers. That is, corporate dollars are falling out of their pockets. A new report from Nucleus Research takes a close look at BYOD costs and finds that companies are financing the trend with little in return.
"The hard ROI of BYOD is a straightforward accounting exercise that is being confused by the feel-good claims around productivity and vendor proclamations that lack a financial foundation," writes Hyoun Park, principal analyst at Nucleus and author of the report.
Similar to this Article

BYOD, of course, was supposed to save companies money. Let's start with its popular premise: Companies no longer have to buy corporate smartphones and tablets. But this is flawed logic, because the cost of devices themselves make up less than 10 percent of a company's annual mobility spend, says Park.
Consider that an enterprise mobile phone costs $200 per device and has an average lifespan of 18 months, at which point an employee asks for an upgrade. This works out to around $11 per month. Tack on additional 20 to 30 percent savings for volume discounts and free backup devices. This means that companies offloading the device cost to employees are saving only $8 per month under BYOD.
Here's the kicker: BYOD increases the other 90 percent of the mobility spend, which includes voice and data, help desk, developers and mobile management software.
BYOD's Big Spend
A typical business user will spend $80 to $90 per month for a personal smartphone voice and data plan, which the company usually reimburses. The careful reader will notice that this is ten times the cost of the device itself. In comparison, a corporate-owned smartphone costs around $60 to $65 per month, thanks to bulk discounts, pooled data for voice and data and texting, and special rates for international reporting, according to Nucleus.
The BYOD premium doesn't stop there, either. There is a hidden cost to process expense reporting and reimbursement, which works out to around $20. (For more on this, check out BYOD: If You Think You're Saving Money, Think Again.)
"In general, any reimbursement above $40 per month means that the company is deliberately giving up money to support BYOD," says Park, adding, "Companies providing a standard $75 reimbursement (or more) through an expense report process are giving up hundreds of dollars per employee every year to support BYOD."
Unintended Consequences
In its BYOD report, Nucleus cites a startling legal case that took place some six months ago: Massachusetts Eye and Ear Infirmary and Massachusetts Eye and Ear Associates settled with the U.S. Department of Health and Human Services for $1.5 million. What happened? Violations of the HIPAA (Health Insurance Portability and Accountability Act) regulation due to the loss of a personal laptop with identifiable health information.
In another example, some companies embroiled in a class action lawsuit against AT&T Mobility stand to lose out on thousands of dollars because they can't participate in a $153 million settlement, says a source knowledgeable about the lawsuit. The reason is because they moved to BYOD.
There's also growing concern that BYOD will open the floodgates to employees suing their employers. Employees are questioning the intrusion of corporate eyes on their personal devices and wondering if companies are taking advantage of them through BYOD.
Similar to this Article
"I anticipate a bunch of little [lawsuits], then something big will happen that'll be a class action and become headline news," CEO John Marshall at AirWatch, an enterprise mobile device management (MDM) vendor with 6,500 customers, including Lowe's, United Airlines and Best Buy, told CIO.com.
What About Productivity Gains?
Nucleus admits BYOD productivity gains are a moving target—that is, they're hard to calculate. Of course, this hasn't stopped companies from trying to calculate the return on investment. Cisco says its BYOD saves the company $2 million per year.
Also, Intel claims to be saving 57 minutes a day for 23,500 BYOD employees. Since the chip giant makes about $500,000 per employee in a year, an hour savings a day per employee works out to around $700 million in added productivity.
"This is a difficult estimate to believe," says Park. "However, if true, a $700 million productivity increase is material to the business and should be considered by the investment community as a key differentiator."
A case for BYOD productivity can be made using a time-tracking model similar to Intel's. Let's say employees spend a day to set up their corporate-liable device, whereas no time at all (or at least not on company time) to set up a personal device.
Since employees in a profit center are expected to bring in around $250,000 per year, a day lost works out to be $1,000.
"This time lost is difficult to overcome based on the costs and benefits associated with BYOD versus a corporate-owned program," Park says.
Tom Kanshige covers Apple, BYOD and Consumerization of IT for CIO.com