影像技術(shù)已經(jīng)不再只是CMOS影像傳感器之間永無止盡的的像素競賽，隨著市場焦點轉(zhuǎn)向“視覺(Vision)”處理，產(chǎn)業(yè)界的新戰(zhàn)線已經(jīng)來到了處理器能如何快速、精準(zhǔn)地以一種能理解嵌入式系統(tǒng)的方式，擷取、分析并詮釋資料。
簡單來說，“誰在看誰”的概念已經(jīng)出現(xiàn)反轉(zhuǎn)。在嵌入式視覺的世界，關(guān)鍵主體不是想拍出更加照片的你或是攝影師們，為嵌入式系統(tǒng)而開發(fā)的技術(shù)要“看”的對象是你、它們要識別你是誰、分析你的行為，然后處理那些它們認(rèn)為你會需要的資料。
可能在你的認(rèn)知里，這些技術(shù)就是“機器視覺”或是“計算機視覺”；也許沒錯，但坦白說，目前有一些利用嵌入式視覺技術(shù)來執(zhí)行的市場行銷策略，令人頗感不安。當(dāng)然，這類市場營銷手法不至于像是美國國家安全局(NSA)的電子監(jiān)聽那么恐怖，但該種技術(shù)基本上就是有一堆傳感器在監(jiān)視我，目標(biāo)就是要賺我的錢──這實在 讓我覺得毛骨悚然。
我最近在日本東京與一家美國嵌入式視覺技術(shù)開發(fā)商CogniVue 的業(yè)務(wù)發(fā)展副總裁Tom Wilson見面；他告訴我，像CogniVue那樣的視覺處理技術(shù)開發(fā)業(yè)者，目標(biāo)市場不會只是汽車應(yīng)用。以下是Wilson所分享的幾個嵌入式視覺技術(shù)應(yīng)用案例：
●當(dāng)駕駛?cè)嗽诤诎祷臎龅牡缆飞祥_車，(平常是關(guān)閉的)路燈會在汽車行進到前方時開啟，當(dāng)它們感應(yīng)到汽車已經(jīng)離開就會再度關(guān)閉。
●當(dāng)你從一個數(shù)字看板──也就是公共場所的大型電子顯示器──前方走過，那個屏幕甚至?xí)谀阕⒁獾剿埃涂梢员鎰e出你的性別與年齡，然后快速變換所顯示的廣告消息，以迎合你所屬的人口族群喜好、吸引你的目光。
●智能手機可識別你的手勢，或者是支持人臉識別協(xié)助你標(biāo)記影像(通知你那個你正在看的人是誰、或是你所拍的照片里有那些人，還會上傳到社群網(wǎng)站)。
●有一雙“眼睛”的機頂盒，會觀察你家客廳、識別出誰在看哪些電視節(jié)目，然后將那些信息傳送到后端服務(wù)器，在你正在看的節(jié)目里置入你會有興趣的數(shù)字產(chǎn)品廣告。
在以上幾個案例里面，最吸引我注意的就是最后那種“有眼睛的機頂盒”；當(dāng)然，因為知道微軟(Microsoft) Xbox 360游戲機的體感識別裝置 Kinect ，這種技術(shù)或許不那么令人驚訝，但我就是很想進一步了解它的運作原理。
對此Wilson的解釋是：“那就是說，如果你正在看美劇《Friends》，那臺機頂盒會知道是你在看，然后知道你喜歡的是可口可樂而不是百事可樂?！庇谑呛蠖说姆?wù)器能以數(shù)字化的方式，把影集中人物正在使用的植入性營銷產(chǎn)品，換成你喜歡的那種。
Wilson 指出，有一家廣告平臺開發(fā)商Mirriad，就是專門提供這樣的解決方案：“他們的方案就是將置入性廣告類型與觀眾的喜好搭配?！备鶕?jù)他的說法，該種“有 眼睛的機頂盒”并不是一個牽強的概念，Mirriad這家公司最近已經(jīng)與機頂盒供貨商Pace簽署合作協(xié)議，要在英國試用這種解決方案。
本文授權(quán)編譯自EE Times，版權(quán)所有，謝絕轉(zhuǎn)載
第2頁：汽車應(yīng)用是各種視覺處理器的主戰(zhàn)場
第3頁：仍有待克服的市場障礙
相關(guān)閱讀：
• DLP嵌入式平臺打開機器視覺開發(fā)大門
• 你的智能手機或許正用攝像頭監(jiān)控你……
• [圖文報道]讓人眼花繚亂的嵌入式視覺應(yīng)用JbDesmc
{pagination}
在解釋何謂數(shù)字廣告植入性營銷方案時，Wilson開玩笑地表示，這就是他家沒有電視的原因之一；但他也讓我理解了那些嵌入式視覺應(yīng)用帶來的深遠影響，以及嵌入式視覺IP (包括軟件與硬件)供貨商之間的競爭，在近幾年來有越演越烈的趨勢。
CogniVue、 Mobileye、 CEVA 與 Tensilica (現(xiàn)已收歸Cadence 旗下)是目前市場上可提供嵌入式視覺技術(shù)的幾家IP供貨商，Imagination Technology 最近也藉由發(fā)表PowerVR Raptor ISP (image signal processing，影像信號處理)技術(shù)成為該領(lǐng)域的新競爭者。
其它芯片大廠包括Freescale、TI與ST也有推出特殊應(yīng)用視覺處理器產(chǎn)品，但通常是與業(yè)界伙伴合作，或是與嵌入式視覺IP供貨商簽屬授權(quán)協(xié)議。
目 前汽車應(yīng)用是各種視覺處理器的主戰(zhàn)場，因為嵌入式視覺在先進駕駛?cè)溯o助系統(tǒng)(ADAS)內(nèi)扮演要角；汽車廠商正指望ADAS帶來新商機，大力宣傳該系統(tǒng)可 提供的各種安全功能如車道偏離警告、撞擊緩解(collision mitigation)、自動停車，以及盲點提醒等等。
根據(jù)市場研究機構(gòu)IHS的估計，特殊應(yīng)用視覺處理器在汽車市場的應(yīng)用規(guī)模，2013年可達到1.51億美元；該數(shù)字在 2012年為1.37億美元，在2011年則為1.26億美元。
JbDesmc
本文授權(quán)編譯自EE Times，版權(quán)所有，謝絕轉(zhuǎn)載
第3頁：仍有待克服的市場障礙
相關(guān)閱讀：
• DLP嵌入式平臺打開機器視覺開發(fā)大門
• 你的智能手機或許正用攝像頭監(jiān)控你……
• [圖文報道]讓人眼花繚亂的嵌入式視覺應(yīng)用JbDesmc
{pagination}
仍有待克服的市場障礙
不 過目前產(chǎn)業(yè)界其實仍只看到嵌入式視覺的表面；如嵌入式視覺聯(lián)盟(Embedded Vision Alliance)創(chuàng)辦人Jeff Bier先前接受EETimes 美國版訪問時所言：“視覺處理仍有許多非常困難的問題有待解決，就算人們花費大量的時間開發(fā)了一系列嵌入式視覺算法?！?
CogniVue 的Wilson也同意以上看法，他指出，要處理大量的實時數(shù)據(jù)需要非常密集的運算性能，而要以一個強健的方式架構(gòu)出“3D傳感器映像圖(3D sensor map)”，特別是在訴求低功耗的消費性電子裝置中，更是艱難任務(wù)。Wilson解釋，所謂的”3D傳感器映像圖”是解決目前2D計算機視覺基本限制的關(guān)鍵。
舉例來說，2D技術(shù)在影像分割(segmentation，也就是分開背景與前景)、照度(illumination， 支持人臉識別)、相對定位(relative position，辨別畫面中物體相對位置)，以及遮蔽(occlusion，識別人臉前方的手)等方面有問題，而不同3D感測方案都面臨性能上的折衷。 Wilson表示，CogniVue現(xiàn)在正試圖透過算法解決映像圖問題，以催生低成本3D傳感器視覺方案。
對系統(tǒng)設(shè)計工程師來說，要設(shè)計出能有效執(zhí)行不同視覺算法的硬件，是很大的挑戰(zhàn)；那些正在尋找影像/視頻處理解決方案的系統(tǒng)供貨商，可選擇把所有任務(wù)留在CPU里面、將影像任務(wù)交給GPU，或是添加專門處理影像的硬件邏輯。
隨著像Imagination這樣的主流GPU核心供貨商涉足視覺市場，可預(yù)見相關(guān)IP供貨商與芯片廠商的競爭將更加激烈。而我們也可以預(yù)期，未來將會有各種各樣讓人驚艷的嵌入式視覺解決方案出現(xiàn)在日常生活中…拭目以待那個“美麗新世界“吧！
本文授權(quán)編譯自EE Times，版權(quán)所有，謝絕轉(zhuǎn)載
編譯：Judith Cheng
參考英文原文：Embedded Vision: Who's Watching Whom & Why，by Junko Yoshida
相關(guān)閱讀：
• DLP嵌入式平臺打開機器視覺開發(fā)大門
• 你的智能手機或許正用攝像頭監(jiān)控你……
• [圖文報道]讓人眼花繚亂的嵌入式視覺應(yīng)用JbDesmc
{pagination}
Embedded Vision: Who's Watching Whom & Why
Junko Yoshida, Chief International Correspondent
TOKYO — Imaging technology is no longer just about the never-ending megapixel race among CMOS image sensors. As market focus shifts to "vision" processing, the industry has drawn a new battle line -- over how fast and how accurately a processor can capture, dissect, and interpret data in a manner comprehensible to an embedded system.
In short, the whole concept of who's watching whom has flipped.
In the embedded vision world, what matters is not so much you, the photographer, who wants to take better photos; instead, the technology now exists to cater to embedded systems that need to watch you, recognize who you are, analyze your behavior, and process data they think you need.
You might call this just the plain reality of technology progress in machine vision or computer vision. Maybe so. But I confess that some of the embedded vision plots hatched by marketers today are disturbing enough to make me cringe.
None of this stuff, of course, is more worrisome than the NSA's electronic spying programs. But the very notion of a bunch of sensors physically watching me -- solely to make a commercial gain at my expense -- gives me, at least, a slight case of the willies. At worst, it's a reminder of the increasingly Orwellian society we already live in.
Over a cup of coffee in Tokyo, I recently sat down with Tom Wilson, vice president of business development at CogniVue, a Quebec-based embedded vision technology developer. Wilson tried to convince me that automotive isn't the only market being targeted by vision processing technology developers like CogniVue.
Here are a few examples he shared with me -- in terms of what comes next with embedded vision:
? Drive a car on a deserted road in the dark. Street lamps -- normally switched off -- light up the road just in front of your car, as you move forward. As soon as they sense your car is leaving, they go off. (Yeah, I know: an evening's drive through The Twilight Zone.)
? Walk in front of a digital sign -- a gigantic electronic display in a public space. The sign, even before you notice it, recognizes your gender and age, then quickly changes the ad message -- to fit your demographic profile -- as you look at it. (Yeah, I know: shades of Minority Report.)
? Smartphones that can recognize your hand gestures, or that can do face recognitions to help you tag images (by informing you who you are seeing, and whose pictures you are taking, and even uploading to social networks.)
? A set-top box embedded with eyes in your living room identifies who is watching what program. It sends the information to a backend server, triggering a digital product placement in a TV program. (Right. Saw that in Fahrenheit 451.)
Among these examples, what ticked me off was the last item about a set-top box with eyes. Of course, for someone who's known Kinect (a motion sensing input device by Microsoft for the Xbox 360 video game console and Windows PCs), I probably shouldn't have been so surprised. But I needed further clarification over what it exactly does.
"Say you are watching Friends. The set-top box knows you're watching it and you actually like Pepsi instead of Coke," explained CogniVue’s Wilson. The backend server, then, can digitally insert a Pepsi can, replacing a Coke, in Monica's living room.
Click here to watch Mirriad's video explaining how its services work.
(Source: Mirrad)
Wilson pointed out that Mirriad, a developer of ad platforms, is one company working on such a project. "The plan is to couple this type of ad insertion with viewer preference," he explained. In fact, a set-top box with eyes isn't such a far-fetched idea. Mirriad recently signed a deal with Pace, a set-top box vendor, to trial this in the UK, according to Wilson.
While explaining the digital product placement scheme, Wilson joked that this is partly why he doesn't own a TV. But he made sure that I understood the far-reaching ramifications of embedded vision applications and how the competition among embedded vision IP vendors -- both software and hardware -- has been escalating in recent years.
CogniVue, Mobileye, CEVA, and Tensilica (now a part of Cadence) are just a few examples of IP companies enabling embedded vision technologies. The newest member to join the fray is Imagination Technology, which announced its PowerVR Raptor ISP (image signal processing) architecture Monday.
Leading chip companies such as Freescale, Texas Instruments, and STMicroelectronics are also rolling out purpose-built vision processors -- often taking advantage of their partnership/licensing deals with embedded vision IP vendors.
For the time being, though, automotive is the primary market for all these vision processors, since embedded vision is playing a key role in Advanced Driver Assist System (ADAS). Carmakers are banking on ADAS, advocating safety features such as lane departure warnings, collision mitigation, self-parking, and blind-spot notification.
According to IHS, a market research firm, revenue in 2013 for special-purpose computer vision processors used in under-the-hood automotive applications is forecast to reach $151 million, up from $137 million last year and from $126 million in 2011.
Hard problems to solve
I should, however, note that the industry is still scratching only the surface of the embedded vision future.
"Vision processing still remains as a very hard problem to solve,"Jeff Bier, founder of the Embedded Vision Alliance, once told EE Times, "despite the number of man-years spent developing a host of embedded vision algorithms."
CogniVue’s Wilson agreed. Processing a huge amount of real-time data demands intense compute power. To do a 3D sensor map in a robust manner, especially in a low-power consumer device, is especially tough, he added.
Asked why a 3D sensor map, he described it as "essential" to solve fundamental limitations in 2D computer vision. He noted that 2D, for example, has problems with segmentation (separating foreground from background), illumination (for face recognition), relative position (placing objects in the scene), and occlusion (hands in front of the face). Noting that different approaches for 3D sensing are fraught with tradeoffs, Wilson said that CogniVue is currently working on an algorithmic way to efficiently compute disparity maps for low-cost 3D sensor vision.
Designing hardware that can efficiently run different vision algorithms is a huge challenge for system designers. Options for system vendors looking for imaging/video processing solutions range from keeping it all in the CPU to offloading imaging to the GPU, or adding hardwired logic dedicated to imaging functions.
With the world's GPU IP core leader Imagination entering the vision market, the race among IP vendors and chip suppliers has only gotten even more intense.
There is no question that it's going to be a "Brave New World."
